怎么读这张表
汇总这些价格信息，帮助大家合理选择适合的模型！
价格单位是美元 / 每百万 token，标准档——不含 batch（通常五折）、缓存命中、快速模式这些另计的档位。
「官方」一列是人工核对的标价，也是折扣标签的判定基准。之所以不自动抓：聚合平台只反映当前生效价，不区分标价和促销价。比如 Claude Sonnet 5 官方标价 $3/$15，但现在各平台都报 $2/$10——那是 2026-08-31 到期的限时优惠，到期后会悄悄变回去，而页面不会有任何提示。
点击表头可以按价格升序 / 降序排列，上方输入框可以按名称筛选。
| 模型 | 官方标价USD / 百万 token | OpenRouterUSD / 百万 token | ZenMuxUSD / 百万 token | 优惠相对官方标价 | 订阅覆盖包含该模型的套餐 | 上下文 | 能力简介 |
|---|---|---|---|---|---|---|---|
| Mistral: Mistral Nemo mistralai/mistral-nemo | 0.15/0.15 | 0.019/0.03 | 未上架 | openrouter 低 87% | ZenMux Ultra | 131K | A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA.… |
| Qwen: Qwen3.7 Flash qwen/qwen3.7-flash | 0.03/0.13 ~ | 0.03/0.13 | 0.03/0.13 | — | OpenCode Go ZenMux Ultra | 1000K | Qwen3.7 Flash is a vision-language reasoning model from Alibaba.… |
| OpenAI: gpt-oss-120b openai/gpt-oss-120b | 0.039/0.18 | 0.03/0.17 | 未上架 | openrouter 低 23% | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 131K | gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-p… |
| OpenAI: gpt-oss-20b openai/gpt-oss-20b | 0.029/0.14 | 0.03/0.13 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 131K | gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.… |
| Qwen: Qwen3 30B A3B Instruct 2507 qwen/qwen3-30b-a3b-instruct-2507 | 0.13/0.52 ~ | 0.0481/0.193 | 未上架 | openrouter 低 63% | OpenCode Go ZenMux Ultra | 262K | Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference.… |
| OpenAI: GPT-5 Nano openai/gpt-5-nano | 0.05/0.4 | 0.05/0.4 | 0.05/0.4 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | GPT-5-Nano is the smallest and fastest variant in the GPT-5 system, optimized for developer tools, rapid interactions, and ultra-low latency environme… |
| Mistral: Mistral Small 3 mistralai/mistral-small-24b-instruct-2501 | 0.05/0.08 | 0.05/0.08 | 未上架 | — | ZenMux Ultra | 33K | Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks.… |
| Qwen: Qwen3.5-Flash qwen/qwen3.5-flash-02-23 | 0.065/0.26 ~ | 0.065/0.26 | 未上架 | — | OpenCode Go ZenMux Ultra | 1000K | The Qwen3.5 native vision-language Flash models are built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-… |
| Google: Gemma 4 26B A4B google/gemma-4-26b-a4b-it | 0.06/0.33 | 0.07/0.34 | 0.15/0.6 | — | ZenMux Ultra | 262K | Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind.… |
| Qwen: Qwen3 Coder 30B A3B Instruct qwen/qwen3-coder-30b-a3b-instruct | 0.2925/1.4625 ~ | 0.07/0.28 | 未上架 | openrouter 低 76% | OpenCode Go ZenMux Ultra | 262K | Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced c… |
| OpenAI: gpt-oss-safeguard-20b openai/gpt-oss-safeguard-20b | 0.075/0.3 | 0.075/0.3 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 131K | gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b.… |
| DeepSeek: DeepSeek V4 Flash 0423 deepseek/deepseek-v4-flash | 0.14/0.28 | 0.0826/0.1652 | 0.22/0.66 | openrouter 低 41% | OpenCode Go ZenMux Ultra | 1049K | DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supportin… |
| Qwen: Qwen3 235B A22B Instruct 2507 qwen/qwen3-235b-a22b-2507 | 0.1495/0.598 ~ | 0.09/0.55 | 0.28/1.11 | openrouter 低 40% | OpenCode Go ZenMux Ultra | 262K | Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B ac… |
| Mistral: Mistral Small 3.2 24B mistralai/mistral-small-3.2-24b-instruct | 0.075/0.2 | 0.0938/0.25 | 未上架 | — | ZenMux Ultra | 256K | Mistral-Small-3.2-24B-Instruct-2506 is an updated 24B parameter model from Mistral optimized for instruction following, repetition reduction, and impr… |
| Google: Gemma 4 31B google/gemma-4-31b-it | 0.12/0.36 | 0.1/0.34 | 未上架 | openrouter 低 17% | ZenMux Ultra | 262K | Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output.… |
| Mistral: Ministral 3 3B 2512 mistralai/ministral-3b-2512 | 0.1/0.1 | 0.1/0.1 | 未上架 | — | ZenMux Ultra | 131K | The smallest model in the Ministral 3 family, Ministral 3 3B is a powerful, efficient tiny language model with vision capabilities. |
| Mistral: Voxtral Small 24B 2507 mistralai/voxtral-small-24b-2507 | 0.1/0.3 | 0.1/0.3 | 未上架 | — | ZenMux Ultra | 32K | Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text perform… |
| Qwen: Qwen3 Next 80B A3B Instruct qwen/qwen3-next-80b-a3b-instruct | 0.0975/0.78 ~ | 0.1/1.1 | 未上架 | — | OpenCode Go ZenMux Ultra | 262K | Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces… |
| Google: Gemini 2.5 Flash Lite google/gemini-2.5-flash-lite | 0.1/0.4 | 0.1/0.4 | 0.1/0.4 | — | ZenMux Ultra | 1049K | Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency.… |
| OpenAI: GPT-4.1 Nano openai/gpt-4.1-nano | 0.1/0.4 | 0.1/0.4 | 0.1/0.4 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 1048K | For tasks that demand low latency, GPT‑4.1 nano is the fastest and cheapest model in the GPT-4.1 series.… |
| Qwen: Qwen3 VL 32B Instruct qwen/qwen3-vl-32b-instruct | 0.104/0.416 ~ | 0.104/0.416 | 未上架 | — | OpenCode Go ZenMux Ultra | 131K | Qwen3-VL-32B-Instruct is a large-scale multimodal vision-language model designed for high-precision understanding and reasoning across text, images, a… |
| Qwen: Qwen3 VL 8B Instruct qwen/qwen3-vl-8b-instruct | 0.117/0.455 ~ | 0.117/0.455 | 未上架 | — | OpenCode Go ZenMux Ultra | 262K | Qwen3-VL-8B-Instruct is a multimodal vision-language model from the Qwen3-VL series, built for high-fidelity understanding and reasoning across text, … |
| Qwen: Qwen3 8B qwen/qwen3-8b | 0.117/0.455 ~ | 0.117/0.455 | 未上架 | — | OpenCode Go ZenMux Ultra | 131K | Qwen3-8B is a dense 8.2B parameter causal language model from the Qwen3 series, designed for both reasoning-heavy tasks and efficient dialogue.… |
| Qwen: Qwen3 Coder Next qwen/qwen3-coder-next | 0.3/1.5 ~ | 0.12/0.8 | 未上架 | openrouter 低 60% | OpenCode Go ZenMux Ultra | 262K | Qwen3-Coder-Next is an open-weight causal language model optimized for coding agents and local development workflows.… |
| Qwen: Qwen3 14B qwen/qwen3-14b | 0.2275/0.91 ~ | 0.12/0.24 | 0.14/1.4 | openrouter 低 47% zenmux 低 38% | OpenCode Go ZenMux Ultra | 131K | Qwen3-14B is a dense 14.8B parameter causal language model from the Qwen3 series, designed for both complex reasoning and efficient dialogue.… |
| Qwen: Qwen3 VL 30B A3B Instruct qwen/qwen3-vl-30b-a3b-instruct | 0.13/0.52 ~ | 0.13/0.52 | 未上架 | — | OpenCode Go ZenMux Ultra | 262K | Qwen3-VL-30B-A3B-Instruct is a multimodal model that unifies strong text generation with visual understanding for images and videos.… |
| Qwen: Qwen3 30B A3B qwen/qwen3-30b-a3b | 0.13/0.52 ~ | 0.13/0.52 | 未上架 | — | OpenCode Go ZenMux Ultra | 131K | Qwen3, the latest generation in the Qwen large language model series, features both dense and mixture-of-experts (MoE) architectures to excel in reaso… |
| DeepSeek: DeepSeek V4 Flash 0731 deepseek/deepseek-v4-flash-0731 | 0.14/0.28 | 0.14/0.28 | 未上架 | — | OpenCode Go ZenMux Ultra | 1311K | DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total.… |
| Mistral: Mistral Small 4 mistralai/mistral-small-2603 | 0.15/0.6 | 0.15/0.6 | 未上架 | — | ZenMux Ultra | 262K | Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single syst… |
| Mistral: Ministral 3 8B 2512 mistralai/ministral-8b-2512 | 0.1/1 | 0.15/0.15 | 未上架 | — | ZenMux Ultra | 262K | A balanced model in the Ministral 3 family, Ministral 3 8B is a powerful, efficient tiny language model with vision capabilities. |
| Qwen: Qwen3 Next 80B A3B Thinking qwen/qwen3-next-80b-a3b-thinking | 0.15/1.2 ~ | 0.15/1.2 | 未上架 | — | OpenCode Go ZenMux Ultra | 262K | Qwen3-Next-80B-A3B-Thinking is a reasoning-first chat model in the Qwen3-Next line that outputs structured “thinking” traces by default.… |
| OpenAI: GPT-4o-mini openai/gpt-4o-mini | 0.15/0.6 | 0.15/0.6 | 0.15/0.6 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 128K | GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs.… |
| OpenAI: GPT-4o-mini (2024-07-18) openai/gpt-4o-mini-2024-07-18 | 0.15/0.6 | 0.15/0.6 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 128K | GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs.… |
| Qwen: Qwen3 VL 8B Thinking qwen/qwen3-vl-8b-thinking | 0.18/2.1 ~ | 0.18/2.1 | 未上架 | — | OpenCode Go ZenMux Ultra | 131K | Qwen3-VL-8B-Thinking is the reasoning-optimized variant of the Qwen3-VL-8B multimodal model, designed for advanced visual and textual reasoning across… |
| Qwen: Qwen3.6 Flash qwen/qwen3.6-flash | 0.1875/1.125 ~ | 0.1875/1.125 | 0.25/1.5 | — | OpenCode Go ZenMux Ultra | 1000K | Qwen3.6 Flash is a fast, efficient language model from Alibaba's Qwen 3.6 series.… |
| Qwen: Qwen3.5-27B qwen/qwen3.5-27b | 0.195/1.56 ~ | 0.195/1.56 | 未上架 | — | OpenCode Go ZenMux Ultra | 262K | The Qwen3.5 27B native vision-language Dense model incorporates a linear attention mechanism, delivering fast response times while balancing inference… |
| Qwen: Qwen3 Coder Flash qwen/qwen3-coder-flash | 0.195/0.975 ~ | 0.195/0.975 | 未上架 | — | OpenCode Go ZenMux Ultra | 1000K | Qwen3 Coder Flash is Alibaba's fast and cost efficient version of their proprietary Qwen3 Coder Plus.… |
| OpenAI: GPT-5.6 Luna Pro openai/gpt-5.6-luna-pro | 0.2/1.2 ~ | 0.2/1.2 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 1050K | GPT-5.6 Luna Pro is the same underlying model as [GPT-5.6 Luna](https://openrouter.ai/openai/gpt-5.6-luna), served with `reasoning.mode` set to `pro` … |
| OpenAI: GPT-5.6 Luna openai/gpt-5.6-luna | 1/6 >272K: 2 | 0.2/1.2 | 0.2/1.2 | openrouter 低 80% zenmux 低 80% | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 1050K | GPT-5.6 Luna is a fast, cost-efficient model in OpenAI's GPT-5.6 series.… |
| OpenAI: GPT-5.4 Nano openai/gpt-5.4-nano | 0.2/1.25 | 0.2/1.25 | 0.2/1.25 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | GPT-5.4 nano is the most lightweight and cost-efficient variant of the GPT-5.4 family, optimized for speed-critical and high-volume tasks.… |
| Mistral: Ministral 3 14B 2512 mistralai/ministral-14b-2512 | 0.2/0.2 | 0.2/0.2 | 未上架 | — | ZenMux Ultra | 262K | The largest model in the Ministral 3 family, Ministral 3 14B offers frontier capabilities and performance comparable to its larger Mistral Small 3.2 2… |
| Qwen: Qwen3 VL 30B A3B Thinking qwen/qwen3-vl-30b-a3b-thinking | 0.2/2.4 ~ | 0.2/2.4 | 未上架 | — | OpenCode Go ZenMux Ultra | 262K | Qwen3-VL-30B-A3B-Thinking is a multimodal model that unifies strong text generation with visual understanding for images and videos.… |
| Qwen: Qwen3 30B A3B Thinking 2507 qwen/qwen3-30b-a3b-thinking-2507 | 0.2/2.4 ~ | 0.2/2.4 | 未上架 | — | OpenCode Go ZenMux Ultra | 82K | Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking.… |
| Mistral: Saba mistralai/mistral-saba | 0.2/0.6 | 0.2/0.6 | 未上架 | — | ZenMux Ultra | 33K | Mistral Saba is a 24B-parameter language model specifically designed for the Middle East and South Asia, delivering accurate and contextually relevant… |
| Qwen: Qwen3 VL 235B A22B Instruct qwen/qwen3-vl-235b-a22b-instruct | 0.26/1.04 ~ | 0.21/1.9 | 未上架 | openrouter 低 19% | OpenCode Go ZenMux Ultra | 262K | Qwen3-VL-235B-A22B Instruct is an open-weight multimodal model that unifies strong text generation with visual understanding across images and video.… |
| Qwen: Qwen3.5-35B-A3B qwen/qwen3.5-35b-a3b | 0.1625/1.3 ~ | 0.225/1.8 | 未上架 | — | OpenCode Go ZenMux Ultra | 262K | The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a spa… |
| Qwen: Qwen3 235B A22B Thinking 2507 qwen/qwen3-235b-a22b-thinking-2507 | 0.23/2.3 ~ | 0.23/2.3 | 0.28/2.78 | — | OpenCode Go ZenMux Ultra | 262K | Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks.… |
| Google: Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) google/gemini-3.1-flash-lite-image | 0.25/1.5 | 0.25/1.5 | 未上架 | — | ZenMux Ultra | 66K | Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) is Google's fastest, most cost-efficient Gemini image model, built for high-velocity developer pipeli… |
| Google: Gemini 3.1 Flash Lite google/gemini-3.1-flash-lite | 0.25/1.5 | 0.25/1.5 | 0.25/1.5 | — | ZenMux Ultra | 1049K | Gemini 3.1 Flash Lite is Google’s GA high-efficiency multimodal model optimized for low-latency, high-volume workloads.… |
| Google: Gemini 3.1 Flash Lite Preview google/gemini-3.1-flash-lite-preview | 0.25/1.5 | 0.25/1.5 | 未上架 | — | ZenMux Ultra | 1049K | Gemini 3.1 Flash Lite Preview is Google's high-efficiency model optimized for high-volume use cases.… |
| OpenAI: GPT-5.1-Codex-Mini openai/gpt-5.1-codex-mini | 0.25/2 | 0.25/2 | 0.25/2 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | GPT-5.1-Codex-Mini is a smaller and faster version of GPT-5.1-Codex |
| OpenAI: GPT-5 Mini openai/gpt-5-mini | 0.25/2 | 0.25/2 | 0.25/2 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | GPT-5 Mini is a compact version of GPT-5, designed to handle lighter-weight reasoning tasks.… |
| Anthropic: Claude 3 Haiku anthropic/claude-3-haiku | 0.25/1.25 | 0.25/1.25 | 未上架 | — | Claude Pro Claude Max ZenMux Ultra | 200K | Claude 3 Haiku is Anthropic's fastest and most compact model for near-instant responsiveness. Quick and accurate targeted performance.… |
| Qwen: Qwen3.5 Plus 2026-02-15 qwen/qwen3.5-plus-02-15 | 0.26/1.56 ~ | 0.26/1.56 | 未上架 | — | OpenCode Go ZenMux Ultra | 1000K | The Qwen3.5 native vision-language series Plus models are built on a hybrid architecture that integrates linear attention mechanisms with sparse mixtu… |
| Qwen: Qwen Plus 0728 qwen/qwen-plus-2025-07-28 | 0.26/0.78 ~ | 0.26/0.78 | 未上架 | — | OpenCode Go ZenMux Ultra | 1000K | Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed, and cost combin… |
| Qwen: Qwen-Plus qwen/qwen-plus | 0.26/0.78 ~ | 0.26/0.78 | 未上架 | — | OpenCode Go ZenMux Ultra | 1000K | Qwen-Plus, based on the Qwen2.5 foundation model, is a 131K context model with a balanced performance, speed, and cost combination. |
| DeepSeek: DeepSeek V3.2 deepseek/deepseek-v3.2 | 0.2288/0.3432 | 0.269/0.4 | 0.293/0.4395 | — | OpenCode Go ZenMux Ultra | 164K | DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance.… |
| DeepSeek: DeepSeek V3.2 Exp deepseek/deepseek-v3.2-exp | 0.27/0.41 | 0.27/0.41 | 0.216/0.328 | zenmux 低 20% | OpenCode Go ZenMux Ultra | 164K | DeepSeek-V3.2-Exp is an experimental large language model released by DeepSeek as an intermediate step between V3.1 and future architectures.… |
| DeepSeek: DeepSeek V3.1 Terminus deepseek/deepseek-v3.1-terminus | 0.27/0.95 | 0.27/1 | 未上架 | — | OpenCode Go ZenMux Ultra | 164K | DeepSeek-V3.1 Terminus is an update to [DeepSeek V3.1](/deepseek/deepseek-chat-v3.1) that maintains the model's original capabilities while addressing… |
| Qwen: Qwen3.5-122B-A10B qwen/qwen3.5-122b-a10b | 0.26/2.08 ~ | 0.29/2.4 | 未上架 | — | OpenCode Go ZenMux Ultra | 262K | The Qwen3.5 122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixtur… |
| Google: Gemini 3.5 Flash Lite google/gemini-3.5-flash-lite | 0.3/2.5 | 0.3/2.5 | 0.3/2.5 | — | ZenMux Ultra | 1049K | Gemini 3.5 Flash Lite is a high-efficiency model from Google with upgraded agentic capabilities.… |
| Qwen: Qwen3.5 Plus 2026-04-20 qwen/qwen3.5-plus-20260420 | 0.3/1.8 ~ | 0.3/1.8 | 未上架 | — | OpenCode Go ZenMux Ultra | 1000K | Qwen3.5 Plus (April 2026) is a large-scale multimodal language model from Alibaba.… |
| Google: Nano Banana (Gemini 2.5 Flash Image) google/gemini-2.5-flash-image | 0.3/30 | 0.3/2.5 | 未上架 | — | ZenMux Ultra | 33K | Gemini 2.5 Flash Image, a.k.a. "Nano Banana," is now generally available.… |
| Mistral: Codestral 2508 mistralai/codestral-2508 | 0.3/0.9 | 0.3/0.9 | 未上架 | — | ZenMux Ultra | 256K | Mistral's cutting-edge language model for coding released end of July 2025.… |
| Qwen: Qwen3 Coder 480B A35B qwen/qwen3-coder | 0.975/4.875 ~ | 0.3/1 | 1.25/5.01 | openrouter 低 69% | OpenCode Go ZenMux Ultra | 262K | Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team.… |
| Google: Gemini 2.5 Flash google/gemini-2.5-flash | 0.3/2.5 | 0.3/2.5 | 0.3/2.5 | — | ZenMux Ultra | 1049K | Gemini 2.5 Flash is Google's state-of-the-art workhorse model, specifically designed for advanced reasoning, coding, mathematics, and scientific tasks… |
| Qwen: Qwen3.7 Plus qwen/qwen3.7-plus | 0.32/1.28 ~ | 0.32/1.28 | 0.4/1.6 | — | OpenCode Go ZenMux Ultra | 1000K | Qwen3.7-Plus is a cost-effective model in Alibaba's Qwen3.7 series.… |
| Qwen: Qwen3.6 Plus qwen/qwen3.6-plus | 0.325/1.95 ~ | 0.325/1.95 | 0.5/3 | — | OpenCode Go ZenMux Ultra | 1000K | Qwen 3.6 Plus builds on a hybrid architecture that combines efficient linear attention with sparse mixture-of-experts routing, enabling strong scalabi… |
| Mistral: Mistral Small 3.1 24B mistralai/mistral-small-3.1-24b-instruct | 0.351/0.555 | 0.351/0.555 | 未上架 | — | ZenMux Ultra | 128K | Mistral Small 3.1 24B Instruct is an upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal capabilities… |
| Google: Gemini 3.7 Flash google/gemini-3.7-flash | 0.375/1.875 ~ | 0.375/1.875 | 0.75/3.75 | — | ZenMux Ultra | 1049K | Gemini 3.7 Flash is a multimodal model from Google for fast agentic workflows, coding, and complex multi-step reasoning.… |
| Qwen: Qwen3.5 397B A17B qwen/qwen3.5-397b-a17b | 0.39/2.34 ~ | 0.39/2.34 | 未上架 | — | OpenCode Go ZenMux Ultra | 262K | The Qwen3.5 series 397B-A17B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse… |
| Qwen: Qwen3 VL 235B A22B Thinking qwen/qwen3-vl-235b-a22b-thinking | 0.4/4 ~ | 0.4/4 | 未上架 | — | OpenCode Go ZenMux Ultra | 131K | Qwen3-VL-235B-A22B Thinking is a multimodal model that unifies strong text generation with visual understanding across images and video.… |
| Mistral: Mistral Medium 3.1 mistralai/mistral-medium-3.1 | 0.4/2 | 0.4/2 | 未上架 | — | ZenMux Ultra | 131K | Mistral Medium 3.1 is an updated version of Mistral Medium 3, which is a high-performance enterprise-grade language model designed to deliver frontier… |
| Mistral: Mistral Medium 3 mistralai/mistral-medium-3 | 0.4/2 | 0.4/2 | 未上架 | — | ZenMux Ultra | 131K | Mistral Medium 3 is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operat… |
| OpenAI: GPT-4.1 Mini openai/gpt-4.1-mini | 0.4/1.6 | 0.4/1.6 | 0.4/1.6 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 1048K | GPT-4.1 Mini is a mid-sized model delivering performance competitive with GPT-4o at substantially lower latency and cost.… |
| Qwen: Qwen3 235B A22B qwen/qwen3-235b-a22b | 0.455/1.82 ~ | 0.455/1.82 | 未上架 | — | OpenCode Go ZenMux Ultra | 131K | Qwen3-235B-A22B is a 235B parameter mixture-of-experts (MoE) model developed by Qwen, activating 22B parameters per forward pass.… |
| Google: Nano Banana 2 (Gemini 3.1 Flash Image) google/gemini-3.1-flash-image | 0.5/3 ~ | 0.5/3 | 未上架 | — | ZenMux Ultra | 131K | Gemini 3.1 Flash Image, a.k.a. "Nano Banana 2," is Google’s latest state of the art image generation and editing model, delivering Pro-level visual qu… |
| Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview) google/gemini-3.1-flash-image-preview | 0.5/60 | 0.5/3 | 未上架 | — | ZenMux Ultra | 66K | Gemini 3.1 Flash Image Preview, a.k.a. "Nano Banana 2," is Google’s latest state of the art image generation and editing model, delivering Pro-level v… |
| Google: Gemini 3 Flash Preview google/gemini-3-flash-preview | 0.5/3 | 0.5/3 | 0.5/3 | — | ZenMux Ultra | 1049K | Gemini 3 Flash Preview is a high speed, high value thinking model designed for agentic workflows, multi turn chat, and coding assistance.… |
| Mistral: Mistral Large 3 2512 mistralai/mistral-large-2512 | 0.5/1.5 | 0.5/1.5 | 0.5/1.5 | — | ZenMux Ultra | 262K | Mistral Large 3 2512 is Mistral’s most capable model to date, featuring a sparse mixture-of-experts architecture with 41B active parameters (675B tota… |
| OpenAI: GPT-3.5 Turbo openai/gpt-3.5-turbo | 0.5/1.5 | 0.5/1.5 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 16K | GPT-3.5 Turbo is OpenAI's fastest model. It can understand and generate natural language or code, and is optimized for chat and traditional completion… |
| MoonshotAI: Kimi K2.5 moonshotai/kimi-k2.5 | 0.6/3 | 0.57/2.85 | 0.58/3.02 | openrouter 低 5% zenmux 低 3% | OpenCode Go ZenMux Ultra Kimi 行板 Kimi 中板 Kimi 小快板 Kimi 快板 | 262K | Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm.… |
| MoonshotAI: Kimi K2 0711 moonshotai/kimi-k2 | 0.57/2.3 | 0.57/2.3 | 未上架 | — | OpenCode Go ZenMux Ultra Kimi 行板 Kimi 中板 Kimi 小快板 Kimi 快板 | 131K | Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 bill… |
| Qwen: Qwen3.6 27B qwen/qwen3.6-27b | 0.45/2.7 ~ | 0.6/3.6 | 未上架 | — | OpenCode Go ZenMux Ultra | 262K | Qwen3.6 27B is a dense 27-billion-parameter language model from the Qwen Team at Alibaba, released in April 2026.… |
| OpenAI: GPT Audio Mini openai/gpt-audio-mini | 0.6/2.4 | 0.6/2.4 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 128K | A cost-efficient version of GPT Audio. The new snapshot features an upgraded decoder for more natural sounding voices and maintains better voice consi… |
| MoonshotAI: Kimi K2 Thinking moonshotai/kimi-k2-thinking | 0.6/2.5 | 0.6/2.5 | 未上架 | — | OpenCode Go ZenMux Ultra Kimi 行板 Kimi 中板 Kimi 小快板 Kimi 快板 | 262K | Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning.… |
| Qwen: Qwen3 Coder Plus qwen/qwen3-coder-plus | 0.65/3.25 ~ | 0.65/3.25 | 1/5 | — | OpenCode Go ZenMux Ultra | 1000K | Qwen3 Coder Plus is Alibaba's proprietary version of the Open Source Qwen3 Coder 480B A35B.… |
| Google: Gemma 2 27B google/gemma-2-27b-it | 0.65/0.65 | 0.65/0.65 | 未上架 | — | ZenMux Ultra | 8K | Gemma 2 27B by Google is an open model built from the same research and technology used to create the [Gemini models](/models?q=gemini).… |
| MoonshotAI: Kimi K2.7 Code moonshotai/kimi-k2.7-code | 0.95/4 | 0.71/3.5 | 0.95/4 | openrouter 低 25% | OpenCode Go ZenMux Ultra Kimi 行板 Kimi 中板 Kimi 小快板 Kimi 快板 | 262K | MoonshotAI: Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over lon… |
| Google: Gemini 3.6 Flash google/gemini-3.6-flash | 1.5/7.5 | 0.75/3.75 | 0.75/3.75 | openrouter 低 50% zenmux 低 50% | ZenMux Ultra | 1049K | Gemini 3.6 Flash is a high-efficiency model from Google for coding, agentic workflows, and web and app development.… |
| OpenAI: GPT-5.4 Mini openai/gpt-5.4-mini | 0.75/4.5 | 0.75/4.5 | 0.75/4.5 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | GPT-5.4 mini brings the core capabilities of GPT-5.4 to a faster, more efficient model optimized for high-throughput workloads.… |
| Qwen: Qwen3 Max Thinking qwen/qwen3-max-thinking | 0.78/3.9 ~ | 0.78/3.9 | 未上架 | — | OpenCode Go ZenMux Ultra | 262K | Qwen3-Max-Thinking is the flagship reasoning model in the Qwen3 series, designed for high-stakes cognitive tasks that require deep, multi-step reasoni… |
| Qwen: Qwen3 Max qwen/qwen3-max | 0.78/3.9 ~ | 0.78/3.9 | 1.2/6 | — | OpenCode Go ZenMux Ultra | 262K | Qwen3-Max is an updated release built on the Qwen3 series, offering major improvements in reasoning, instruction following, multilingual support, and … |
| MoonshotAI: Kimi K2.6 moonshotai/kimi-k2.6 | 0.95/4 | 0.95/4 | 0.95/4 | — | OpenCode Go ZenMux Ultra Kimi 行板 Kimi 中板 Kimi 小快板 Kimi 快板 | 262K | Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchest… |
| SpaceXAI: Grok Build 0.1 x-ai/grok-build-0.1 | 1/2 | 1/2 | 1/2 | — | ZenMux Ultra | 256K | Grok Build 0.1 is SpaceXAI’s fast coding model trained specifically for agentic software engineering workflows.… |
| Anthropic: Claude Haiku 4.5 anthropic/claude-haiku-4.5 | 1/5 | 1/5 | 1/5 | — | Claude Pro Claude Max ZenMux Ultra | 200K | Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of large…200K 上下文（其余当代模型为 1M） |
| OpenAI: GPT-3.5 Turbo (older v0613) openai/gpt-3.5-turbo-0613 | 1.5/2 | 1/2 | 未上架 | openrouter 低 33% | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 4K | GPT-3.5 Turbo is OpenAI's fastest model. It can understand and generate natural language or code, and is optimized for chat and traditional completion… |
| Qwen: Qwen3.6 Max Preview qwen/qwen3.6-max-preview | 1.027/6.162 ~ | 1.027/6.162 | 1.3/7.8 | — | OpenCode Go ZenMux Ultra | 262K | Qwen3.6-Max-Preview is a proprietary frontier model from Alibaba Cloud built on a sparse mixture-of-experts architecture with approximately 1 trillion… |
| OpenAI: o4 Mini High openai/o4-mini-high | 1.1/4.4 | 1.1/4.4 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 200K | OpenAI o4-mini-high is the same model as [o4-mini](/openai/o4-mini) with reasoning_effort set to high.… |
| OpenAI: o4 Mini openai/o4-mini | 1.1/4.4 | 1.1/4.4 | 1.1/4.4 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 200K | OpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining strong multimodal and agen… |
| OpenAI: o3 Mini High openai/o3-mini-high | 1.1/4.4 | 1.1/4.4 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 200K | OpenAI o3-mini-high is the same model as [o3-mini](/openai/o3-mini) with reasoning_effort set to high.… |
| OpenAI: o3 Mini openai/o3-mini | 1.1/4.4 | 1.1/4.4 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 200K | OpenAI o3-mini is a cost-efficient language model optimized for STEM reasoning tasks, particularly excelling in science, mathematics, and coding.… |
| SpaceXAI: Grok 4.3 x-ai/grok-4.3 | 1.25/2.5 | 1.25/2.5 | 1.25/2.5 | — | ZenMux Ultra | 1000K | Grok 4.3 is a reasoning model from SpaceXAI. It accepts text and image inputs with text output, and is suited for agentic workflows, instruction-follo… |
| SpaceXAI: Grok 4.20 Multi-Agent x-ai/grok-4.20-multi-agent | 2/6 | 1.25/2.5 | 未上架 | openrouter 低 38% | ZenMux Ultra | 2000K | Grok 4.20 Multi-Agent is a variant of SpaceXAI’s Grok 4.20 designed for collaborative, agent-based workflows.… |
| SpaceXAI: Grok 4.20 x-ai/grok-4.20 | 1.25/2.5 | 1.25/2.5 | 未上架 | — | ZenMux Ultra | 2000K | Grok 4.20 is a reasoning model from SpaceXAI with industry-leading speed and agentic tool calling capabilities.… |
| OpenAI: GPT-5.1-Codex-Max openai/gpt-5.1-codex-max | 1.25/10 | 1.25/10 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | GPT-5.1-Codex-Max is OpenAI’s latest agentic coding model, designed for long-running, high-context software development tasks.… |
| OpenAI: GPT-5.1 openai/gpt-5.1 | 1.25/10 | 1.25/10 | 1.25/10 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | GPT-5.1 is the latest frontier-grade model in the GPT-5 series, offering stronger general-purpose reasoning, improved instruction adherence, and a mor… |
| OpenAI: GPT-5.1-Codex openai/gpt-5.1-codex | 1.25/10 | 1.25/10 | 1.25/10 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | GPT-5.1-Codex is a specialized version of GPT-5.1 optimized for software engineering and coding workflows.… |
| OpenAI: GPT-5 openai/gpt-5 | 1.25/10 | 1.25/10 | 1.25/10 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | GPT-5 is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience.… |
| Google: Gemini 2.5 Pro google/gemini-2.5-pro | 1.25/10 >200K: 2.5 | 1.25/10 | 1.25/10 | — | ZenMux Ultra | 1049K | Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks.… |
| Google: Gemini 2.5 Pro Preview 06-05 google/gemini-2.5-pro-preview | 1.25/10 >200K: 2.5 | 1.25/10 | 未上架 | — | ZenMux Ultra | 1049K | Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks.… |
| Google: Gemini 2.5 Pro Preview 05-06 google/gemini-2.5-pro-preview-05-06 | 1.25/10 >200K: 2.5 | 1.25/10 | 未上架 | — | ZenMux Ultra | 1049K | Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks.… |
| DeepSeek: DeepSeek V4 Pro 0813 deepseek/deepseek-v4-pro-0813 | 0.435/0.87 | 1.32/3.96 | 未上架 | — | OpenCode Go ZenMux Ultra | 1049K | DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek. This is the GA release of DeepSeek V4 Pro. |
| DeepSeek: DeepSeek V4 Pro 0423 deepseek/deepseek-v4-pro | 0.435/0.87 | 1.32/3.96 | 0.66/1.98 | — | OpenCode Go ZenMux Ultra | 1049K | DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token… |
| Qwen: Qwen3.7 Max qwen/qwen3.7-max | 1.475/4.425 ~ | 1.475/4.425 | 1.25/3.75 | zenmux 低 15% | OpenCode Go ZenMux Ultra | 1000K | Qwen3.7-Max is the flagship model in Alibaba's Qwen3.7 series.… |
| Google: Gemini 3.5 Flash google/gemini-3.5-flash | 1.5/9 | 1.5/9 | 1.5/9 | — | ZenMux Ultra | 1049K | Gemini 3.5 Flash is Google's high-efficiency multimodal model, bringing near-Pro level coding and reasoning at Flash-tier cost and speed.… |
| Mistral: Mistral Medium 3.5 mistralai/mistral-medium-3-5 | 0.4/2 | 1.5/7.5 | 未上架 | — | ZenMux Ultra | 262K | Mistral Medium 3.5 is a dense 128B instruction-following model from Mistral AI.… |
| OpenAI: GPT-3.5 Turbo Instruct openai/gpt-3.5-turbo-instruct | 1.5/2 | 1.5/2 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 4K | This model is a variant of GPT-3.5 Turbo tuned for instructional prompts and omitting chat-related optimizations. Training data: up to Sep 2021. |
| OpenAI: GPT-5.3-Codex openai/gpt-5.3-codex | 1.75/14 | 1.75/14 | 1.75/14 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | GPT-5.3-Codex is OpenAI’s most advanced agentic coding model, combining the frontier software engineering performance of GPT-5.2-Codex with the broade… |
| OpenAI: GPT-5.2-Codex openai/gpt-5.2-codex | 1.75/14 | 1.75/14 | 1.75/14 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | GPT-5.2-Codex is an upgraded version of GPT-5.1-Codex optimized for software engineering and coding workflows.… |
| OpenAI: GPT-5.2 Chat openai/gpt-5.2-chat | 1.75/14 | 1.75/14 | 1.75/14 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 128K | GPT-5.2 Chat (AKA Instant) is the fast, lightweight member of the 5.2 family, optimized for low-latency chat while retaining strong general intelligen… |
| OpenAI: GPT-5.2 openai/gpt-5.2 | 1.75/14 | 1.75/14 | 1.75/14 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | GPT-5.2 is the latest frontier-grade model in the GPT-5 series, offering stronger agentic and long context perfomance compared to GPT-5.1.… |
| Qwen: Qwen3.8 2.4T A95B qwen/qwen3.8-2.4t-a95b | 2/6 ~ | 2/6 | 未上架 | — | OpenCode Go ZenMux Ultra | 1049K | Qwen3.8 2.4T A95B is an open-weight sparse mixture-of-experts model from Qwen and the open-weight variant of [Qwen3.8 Max](/qwen/qwen3.8-max), with 95… |
| SpaceXAI: Grok 4.6 x-ai/grok-4.6 | 2/6 ~ | 2/6 | 2/6 | — | ZenMux Ultra | 500K | Grok 4.6 is SpaceXAI's smartest model with frontier performance on coding, knowledge work, and STEM. |
| Qwen: Qwen3.8 Max qwen/qwen3.8-max | 2/6 ~ | 2/6 | 1.4/4.2 | zenmux 低 30% | OpenCode Go ZenMux Ultra | 1000K | Qwen3.8 Max is the flagship model in Alibaba's Qwen3.8 series, the general-availability successor to the Qwen3.8 Max Preview.… |
| OpenAI: GPT-5.6 Terra Pro openai/gpt-5.6-terra-pro | 2/12 ~ | 2/12 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 1050K | GPT-5.6 Terra Pro is the same underlying model as [GPT-5.6 Terra](https://openrouter.ai/openai/gpt-5.6-terra), served with `reasoning.mode` set to `pr… |
| OpenAI: GPT-5.6 Terra openai/gpt-5.6-terra | 2.5/15 >272K: 5 | 2/12 | 2/12 | openrouter 低 20% zenmux 低 20% | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 1050K | GPT-5.6 Terra is a balanced model in OpenAI's GPT-5.6 series, positioned between the flagship Sol tier and the cost-efficient Luna tier.… |
| SpaceXAI: Grok 4.5 x-ai/grok-4.5 | 2/6 | 2/6 | 2/6 | — | ZenMux Ultra | 500K | Grok 4.5 is a model from SpaceXAI with frontier performance on coding, knowledge work, and STEM. |
| Anthropic: Claude Sonnet 5 anthropic/claude-sonnet-5 | 3/15 | 2/10 | 2/10 | openrouter 促销 −33% zenmux 促销 −33% 2026-08-31 到期 | Claude Pro Claude Max ZenMux Ultra | 1000K | Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work.…限时 $2/$10，2026-08-31 到期后回到标价 |
| Google: Nano Banana Pro (Gemini 3 Pro Image) google/gemini-3-pro-image | 2/12 ~ | 2/12 | 未上架 | — | ZenMux Ultra | 131K | Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro.… |
| Google: Gemini 3.1 Pro Preview Custom Tools google/gemini-3.1-pro-preview-customtools | 2/12 >200K: 4 | 2/12 | 未上架 | — | ZenMux Ultra | 1049K | Gemini 3.1 Pro Preview Custom Tools is a variant of Gemini 3.1 Pro that improves tool selection behavior by preventing overuse of a general bash tool … |
| Google: Gemini 3.1 Pro Preview google/gemini-3.1-pro-preview | 2/12 >200K: 4 | 2/12 | 2/12 | — | ZenMux Ultra | 1049K | Gemini 3.1 Pro Preview is Google’s frontier reasoning model, delivering enhanced software engineering performance, improved agentic reliability, and m… |
| Google: Nano Banana Pro (Gemini 3 Pro Image Preview) google/gemini-3-pro-image-preview | 2/120 | 2/12 | 未上架 | — | ZenMux Ultra | 66K | Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro.… |
| OpenAI: o3 openai/o3 | 2/8 ~ | 2/8 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 200K | o3 is a well-rounded and powerful model across domains. It sets a new standard for math, science, coding, and visual reasoning tasks.… |
| OpenAI: GPT-4.1 openai/gpt-4.1 | 2/8 | 2/8 | 2/8 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 1048K | GPT-4.1 is a flagship large language model optimized for advanced instruction following, real-world software engineering, and long-context reasoning.… |
| Mistral Large 2407 mistralai/mistral-large-2407 | 2/6 | 2/6 | 未上架 | — | ZenMux Ultra | 131K | This is Mistral AI's flagship model, Mistral Large 2 (version mistral-large-2407).… |
| Mistral: Mixtral 8x22B Instruct mistralai/mixtral-8x22b-instruct | 0.9/0.9 | 2/6 | 未上架 | — | ZenMux Ultra | 66K | Mistral's official instruct fine-tuned version of [Mixtral 8x22B](/models/mistralai/mixtral-8x22b).… |
| Mistral Large mistralai/mistral-large | 2/6 | 2/6 | 未上架 | — | ZenMux Ultra | 128K | This is Mistral AI's flagship model, Mistral Large 2 (version `mistral-large-2407`).… |
| OpenAI: GPT-5.6 Sol Pro openai/gpt-5.6-sol-pro | 2.5/15 ~ | 2.5/15 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 1050K | GPT-5.6 Sol Pro is the same underlying model as [GPT-5.6 Sol](https://openrouter.ai/openai/gpt-5.6-sol), served with `reasoning.mode` set to `pro` for… |
| OpenAI: GPT-5.6 Sol openai/gpt-5.6-sol | 5/30 >272K: 10 | 2.5/15 | 5/30 | openrouter 低 50% | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 1050K | GPT-5.6 Sol is the flagship model in OpenAI's GPT-5.6 series. It is suited for complex reasoning, coding, and agentic workflows, and is particularly s… |
| OpenAI: GPT-5.4 openai/gpt-5.4 | 2.5/15 >272K: 5 | 2.5/15 | 2.5/15 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 1050K | GPT-5.4 is OpenAI’s latest frontier model, unifying the Codex and GPT lines into a single system.… |
| OpenAI: GPT Audio openai/gpt-audio | 2.5/10 | 2.5/10 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 128K | The gpt-audio model is OpenAI's first generally available audio model.… |
| OpenAI: GPT-5 Image Mini openai/gpt-5-image-mini | 2.5/2 | 2.5/2 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | GPT-5 Image Mini combines OpenAI's advanced language capabilities, powered by [GPT-5 Mini](https://openrouter.ai/openai/gpt-5-mini), with GPT Image 1 … |
| OpenAI: GPT-4o (2024-11-20) openai/gpt-4o-2024-11-20 | 2.5/10 | 2.5/10 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 128K | The 2024-11-20 version of GPT-4o offers a leveled-up creative writing ability with more natural, engaging, and tailored writing to improve relevance &… |
| OpenAI: GPT-4o (2024-08-06) openai/gpt-4o-2024-08-06 | 2.5/10 | 2.5/10 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 128K | The 2024-08-06 version of GPT-4o offers improved performance in structured outputs, with the ability to supply a JSON schema in the respone_format.… |
| OpenAI: GPT-4o openai/gpt-4o | 2.5/10 | 2.5/10 | 2.5/10 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 128K | GPT-4o ("o" for "omni") is OpenAI's latest AI model, supporting both text and image inputs with text outputs.… |
| MoonshotAI: Kimi K3 moonshotai/kimi-k3 | 3/15 | 3/15 | 2.7/13.5 | zenmux 低 10% | OpenCode Go ZenMux Ultra Kimi 行板 Kimi 中板 Kimi 小快板 Kimi 快板 | 1049K | Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI.… |
| Anthropic: Claude Sonnet 4.6 anthropic/claude-sonnet-4.6 | 3/15 | 3/15 | 3/15 | — | Claude Pro Claude Max ZenMux Ultra | 1000K | Sonnet 4.6 is Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work.… |
| Anthropic: Claude Sonnet 4.5 anthropic/claude-sonnet-4.5 | 3/15 >200K: 6 | 3/15 | 3/15 | — | Claude Pro Claude Max ZenMux Ultra | 1000K | Claude Sonnet 4.5 is Anthropic’s most advanced Sonnet model to date, optimized for real-world agents and coding workflows.… |
| Anthropic: Claude Sonnet 4 anthropic/claude-sonnet-4 | 3/15 | 3/15 | 3/15 | — | Claude Pro Claude Max ZenMux Ultra | 1000K | Claude Sonnet 4 significantly enhances the capabilities of its predecessor, Sonnet 3.7, excelling in both coding and reasoning tasks with improved pre… |
| OpenAI: GPT-3.5 Turbo 16k openai/gpt-3.5-turbo-16k | 3/4 | 3/4 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 16K | This model offers four times the context length of gpt-3.5-turbo, allowing it to support approximately 20 pages of text in a single request at a highe… |
| Claude Opus 5 anthropic/claude-opus-5 | 5/25 | 5/25 | 5/25 | — | Claude Pro Claude Max ZenMux Ultra | 1000K | Claude Opus 5 is Anthropic’s flagship model for demanding reasoning, coding, and long-horizon agentic work.…旗舰，1M 上下文；快速模式另计 $10/$50 |
| Anthropic: Claude Opus 4.8 anthropic/claude-opus-4.8 | 5/25 | 5/25 | 5/25 | — | Claude Pro Claude Max ZenMux Ultra | 1000K | Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family.… |
| OpenAI: GPT Chat Latest openai/gpt-chat-latest | 5/30 | 5/30 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | GPT Chat Latest points to OpenAI's stable API alias `chat-latest` that always resolves to the latest Instant chat model used in ChatGPT.… |
| OpenAI: GPT-5.5 openai/gpt-5.5 | 5/30 | 5/30 | 5/30 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 1050K | GPT-5.5 is OpenAI’s frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and i… |
| Anthropic: Claude Opus 4.7 anthropic/claude-opus-4.7 | 5/25 | 5/25 | 5/25 | — | Claude Pro Claude Max ZenMux Ultra | 1000K | Opus 4.7 is the next generation of Anthropic's Opus family, built for long-running, asynchronous agents.… |
| Anthropic: Claude Opus 4.6 anthropic/claude-opus-4.6 | 5/25 | 5/25 | 5/25 | — | Claude Pro Claude Max ZenMux Ultra | 1000K | Opus 4.6 is Anthropic’s strongest model for coding and long-running professional tasks.… |
| Anthropic: Claude Opus 4.5 anthropic/claude-opus-4.5 | 5/25 | 5/25 | 5/25 | — | Claude Pro Claude Max ZenMux Ultra | 200K | Claude Opus 4.5 is Anthropic’s frontier reasoning model optimized for complex software engineering, agentic workflows, and long-horizon computer use.… |
| OpenAI: GPT-4o (2024-05-13) openai/gpt-4o-2024-05-13 | 2.5/10 | 5/15 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 128K | GPT-4o ("o" for "omni") is OpenAI's latest AI model, supporting both text and image inputs with text outputs.… |
| OpenAI: GPT-5.4 Image 2 openai/gpt-5.4-image-2 | 8/15 | 8/15 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 272K | [GPT-5.4](https://openrouter.ai/openai/gpt-5.4) Image 2 combines OpenAI's GPT-5.4 model with state-of-the-art image generation capabilities from GPT I… |
| Claude Opus 5 (Fast) anthropic/claude-opus-5-fast | 5/25 | 10/50 | 未上架 | — | Claude Pro Claude Max ZenMux Ultra | 1000K | Fast-mode variant of [Opus 5](/anthropic/claude-opus-5) - identical capabilities with higher output speed at 2x pricing relative to regular Opus 5.… |
| Anthropic: Claude Fable 5 anthropic/claude-fable-5 | 10/50 | 10/50 | 10/50 | — | Claude Pro Claude Max ZenMux Ultra | 1000K | Claude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding.…最强模型，1M 上下文；需 30 天数据保留，ZDR 组织不可用 |
| Anthropic: Claude Opus 4.8 (Fast) anthropic/claude-opus-4.8-fast | 5/25 | 10/50 | 未上架 | — | Claude Pro Claude Max ZenMux Ultra | 1000K | Fast-mode variant of [Opus 4.8](/anthropic/claude-opus-4.8) - identical capabilities with higher output speed at 2x pricing relative to regular Opus 4… |
| OpenAI: GPT-5 Image openai/gpt-5-image | 10/10 | 10/10 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | [GPT-5](https://openrouter.ai/openai/gpt-5) Image combines OpenAI's GPT-5 model with state-of-the-art image generation capabilities.… |
| OpenAI: GPT-4 Turbo openai/gpt-4-turbo | 10/30 | 10/30 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 128K | The latest GPT-4 Turbo model with vision capabilities. Vision requests can now use JSON mode and function calling. Training data: up to December 2023. |
| OpenAI: GPT-4 Turbo Preview openai/gpt-4-turbo-preview | 10/30 | 10/30 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 128K | The preview GPT-4 model with improved instruction following, JSON mode, reproducible outputs, parallel function calling, and more.… |
| OpenAI: GPT-5 Pro openai/gpt-5-pro | 15/120 | 15/120 | 15/120 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | GPT-5 Pro is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience.… |
| Anthropic: Claude Opus 4.1 anthropic/claude-opus-4.1 | 15/75 | 15/75 | / | — | Claude Pro Claude Max ZenMux Ultra | 200K | Claude Opus 4.1 is an updated version of Anthropic’s flagship model, offering improved performance in coding, reasoning, and agentic tasks.… |
| Anthropic: Claude Opus 4 anthropic/claude-opus-4 | 15/75 | 15/75 | 15/75 | — | Claude Pro Claude Max ZenMux Ultra | 200K | Claude Opus 4 is benchmarked as the world’s best coding model, at time of release, bringing sustained performance on complex, long-running tasks and a… |
| OpenAI: o1 openai/o1 | 15/60 | 15/60 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 200K | The latest and strongest model family from OpenAI, o1 is designed to spend more time thinking before responding.… |
| OpenAI: o3 Pro openai/o3-pro | 20/80 | 20/80 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 200K | The o-series of models are trained with reinforcement learning to think before they answer and perform complex reasoning.… |
| OpenAI: GPT-5.2 Pro openai/gpt-5.2-pro | 21/168 | 21/168 | 21/168 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | GPT-5.2 Pro is OpenAI’s most advanced model, offering major improvements in agentic coding and long context performance over GPT-5 Pro.… |
| Anthropic: Claude Opus 4.7 (Fast) anthropic/claude-opus-4.7-fast | 5/25 | 30/150 | 未上架 | — | Claude Pro Claude Max ZenMux Ultra | 1000K | Fast-mode variant of [Opus 4.7](/anthropic/claude-opus-4.7) - identical capabilities with higher output speed at premium 6x pricing.… |
| OpenAI: GPT-5.5 Pro openai/gpt-5.5-pro | 30/180 | 30/180 | 30/180 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 1050K | GPT-5.5 Pro is OpenAI’s high-capability model optimized for deep reasoning and accuracy on complex, high-stakes workloads.… |
| OpenAI: GPT-5.4 Pro openai/gpt-5.4-pro | 30/180 >272K: 60 | 30/180 | 30/180 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 1050K | GPT-5.4 Pro is OpenAI's most advanced model, building on GPT-5.4's unified architecture with enhanced reasoning capabilities for complex, high-stakes … |
| OpenAI: GPT-4 openai/gpt-4 | 30/60 | 30/60 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 8K | OpenAI's flagship model, GPT-4 is a large-scale multimodal language model capable of solving difficult problems with greater accuracy than previous mo… |
| OpenAI: o1-pro openai/o1-pro | 150/600 | 150/600 | 未上架 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 200K | The o1 series of models are trained with reinforcement learning to think before they answer and perform complex reasoning.… |
| DeepSeek: DeepSeek V4 Flash 0731 deepseek/deepseek-v4-flash-free | 0.14/0.28 | — | / | — | OpenCode Go ZenMux Ultra | 1000K | |
| Google: Gemini Embedding 2 google/gemini-embedding-2 | 0.2/ ~ | — | 0.2/ | — | ZenMux Ultra | 8K | |
| OpenAI: Chat Latest (GPT-5.5 Instant) openai/chat-latest | 5/30 ~ | — | 5/30 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | |
| OpenAI: GPT-Image-2 openai/gpt-image-2 | 5/ | — | 5/ | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 10K | |
| OpenAI: GPT-5.3 Chat openai/gpt-5.3-chat | 1.75/14 | — | 1.75/14 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 128K | |
| OpenAI: GPT-Image-1.5 openai/gpt-image-1.5 | 5/10 | — | 5/10 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 10K | |
| OpenAI: GPT-5.1 Chat openai/gpt-5.1-chat | 1.25/10 | — | 1.25/10 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 128K | |
| OpenAI: Text Embedding 3 Small openai/text-embedding-3-small | 0.02/ | — | 0.02/ | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 8K | |
| OpenAI: Text Embedding 3 Large openai/text-embedding-3-large | 0.13/ | — | 0.13/ | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 8K | |
| OpenAI: GPT-5 Codex openai/gpt-5-codex | 1.25/10 | — | 1.25/10 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 400K | |
| OpenAI: GPT-5 Chat openai/gpt-5-chat | 1.25/10 | — | 1.25/10 | — | ChatGPT Plus ChatGPT Pro ZenMux Ultra | 128K | |
| inclusionAI: Ling-2.6-flash inclusionai/ling-2.6-flash | — | 0.01/0.03 | / | — | — | 262K | Ling-2.6-flash is an instant (instruct) model from inclusionAI with 104B total parameters and 7.4B active parameters, designed for real-world agents t… |
| Ling-3.0-flash inclusionai/ling-3.0-flash | — | 0.021/0.063 | 0.021/0.063 | — | — | 262K | *Ling-3.0-flash* is a *124B-parameter Mixture-of-Experts (MoE) model*, with approximately *5.1B parameters activated per token*.… |
| Meta: Llama 3.2 1B Instruct meta-llama/llama-3.2-1b-instruct | — | 0.027/0.201 | 未上架 | — | ZenMux Ultra | 60K | Llama 3.2 1B is a 1-billion-parameter language model focused on efficiently performing natural language tasks, such as summarization, dialogue, and mu… |
| Amazon: Nova Micro 1.0 amazon/nova-micro-v1 | — | 0.035/0.14 | 未上架 | — | — | 128K | Amazon Nova Micro 1.0 is a text-only model that delivers the lowest latency responses in the Amazon Nova family of models at a very low cost.… |
| Cohere: Command R7B (12-2024) cohere/command-r7b-12-2024 | — | 0.0375/0.15 | 未上架 | — | — | 128K | Command R7B (12-2024) is a small, fast update of the Command R+ model, delivered in December 2024.… |
| NVIDIA: Nemotron 3 Nano 30B A3B nvidia/nemotron-3-nano-30b-a3b | — | 0.05/0.2 | 未上架 | — | — | 262K | NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic … |
| Google: Gemma 3 4B google/gemma-3-4b-it | — | 0.05/0.1 | 未上架 | — | ZenMux Ultra | 131K | Gemma 3 introduces multimodality, supporting vision-language input and text outputs.… |
| Google: Gemma 3 12B google/gemma-3-12b-it | — | 0.05/0.15 | 未上架 | — | ZenMux Ultra | 131K | Gemma 3 introduces multimodality, supporting vision-language input and text outputs.… |
| Meta: Llama 3.2 3B Instruct meta-llama/llama-3.2-3b-instruct | — | 0.05/0.33 | 未上架 | — | ZenMux Ultra | 131K | Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue genera… |
| Meta: Llama 3.1 8B Instruct meta-llama/llama-3.1-8b-instruct | — | 0.05/0.08 | 未上架 | — | ZenMux Ultra | 131K | Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This 8B instruct-tuned version is fast and efficient.… |
| Z.ai: GLM 4.7 Flash z-ai/glm-4.7-flash | — | 0.06/0.4 | 未上架 | — | OpenCode Go ZenMux Ultra GLM Coding | 203K | As a 30B-class SOTA model, GLM-4.7-Flash offers a new option that balances performance and efficiency.… |
| Google: Gemma 3n 4B google/gemma-3n-e4b-it | — | 0.06/0.12 | 未上架 | — | ZenMux Ultra | 33K | Gemma 3n E4B-it is optimized for efficient execution on mobile and low-resource devices, such as phones, laptops, and tablets.… |
| Amazon: Nova Lite 1.0 amazon/nova-lite-v1 | — | 0.06/0.24 | 未上架 | — | — | 300K | Amazon Nova Lite 1.0 is a very low-cost multimodal model from Amazon that focused on fast processing of image, video, and text inputs to generate text… |
| inclusionAI: Ring-2.6-1T inclusionai/ring-2.6-1t | — | 0.075/0.625 | 0.3/2.5 | — | — | 262K | Ring-2.6-1T is a 1T-parameter-scale thinking model with 63B active parameters, built for real-world agent workflows that require both strong capabilit… |
| inclusionAI: Ling-2.6-1T inclusionai/ling-2.6-1t | — | 0.075/0.625 | 0.3/2.5 | — | — | 262K | Ling-2.6-1T is an instant (instruct) model from inclusionAI and the company’s trillion-parameter flagship, designed for real-world agents that require… |
| ByteDance Seed: Seed 1.6 Flash bytedance-seed/seed-1.6-flash | — | 0.075/0.3 | 未上架 | — | — | 262K | Seed 1.6 Flash is an ultra-fast multimodal deep thinking model by ByteDance Seed, supporting both text and visual understanding.… |
| NVIDIA: Nemotron 3.5 Lightning nvidia/nemotron-3.5-lightning | — | 0.08/0.2 | 未上架 | — | — | 1000K | NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA, with 3B active parameters out of 30B total.… |
| Qwen: Qwen3 32B qwen/qwen3-32b | — | 0.08/0.28 | 未上架 | — | OpenCode Go ZenMux Ultra | 131K | Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue.… |
| Google: Gemma 3 27B google/gemma-3-27b-it | — | 0.08/0.45 | 未上架 | — | ZenMux Ultra | 262K | Gemma 3 introduces multimodality, supporting vision-language input and text outputs.… |
| NVIDIA: Nemotron 3 Super nvidia/nemotron-3-super-120b-a12b | — | 0.085/0.4 | 未上架 | — | — | 1000K | NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in compl… |
| Qwen: Qwen3.5-9B qwen/qwen3.5-9b | — | 0.1/0.15 | 未上架 | — | OpenCode Go ZenMux Ultra | 262K | Qwen3.5-9B is a multimodal foundation model from the Qwen3.5 family, designed to deliver strong reasoning, coding, and visual understanding in an effi… |
| ByteDance Seed: Seed-2.0-Mini bytedance-seed/seed-2.0-mini | — | 0.1/0.4 | 未上架 | — | — | 262K | Seed-2.0-mini targets latency-sensitive, high-concurrency, and cost-sensitive scenarios, emphasizing fast response and flexible inference deployment.… |
| StepFun: Step 3.5 Flash stepfun/step-3.5-flash | — | 0.1/0.3 | 0.1/0.3 | — | — | 262K | Step 3.5 Flash is StepFun's most capable open-source foundation model.… |
| Meta: Llama 4 Scout meta-llama/llama-4-scout | — | 0.1/0.3 | 未上架 | — | ZenMux Ultra | 1311K | Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 10… |
| Meta: Llama 3.3 70B Instruct meta-llama/llama-3.3-70b-instruct | — | 0.1/0.32 | 未上架 | — | ZenMux Ultra | 131K | The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out).… |
| Qwen: Qwen2.5 7B Instruct qwen/qwen-2.5-7b-instruct | — | 0.1/0.2 | 未上架 | — | OpenCode Go ZenMux Ultra | 33K | Qwen2.5 7B is the latest series of Qwen large language models.… |
| Z.ai: GLM 4.5 Air z-ai/glm-4.5-air | — | 0.13/0.85 | 0.1165/0.2911 | — | OpenCode Go ZenMux Ultra GLM Coding | 131K | GLM-4.5-Air is the lightweight variant of our latest flagship model family, also purpose-built for agent-centric applications.… |
| Tencent: Hy3 tencent/hy3 | — | 0.132/0.528 | 0.147/0.589 | — | — | 262K | Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and… |
| Qwen: Qwen3.6 35B A3B qwen/qwen3.6-35b-a3b | — | 0.14/1 | 未上架 | — | OpenCode Go ZenMux Ultra | 262K | Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token.… |
| Xiaomi: MiMo-V2.5 xiaomi/mimo-v2.5 | — | 0.14/0.28 | 0.12/0.232 | — | OpenCode Go MiMo Lite MiMo Standard MiMo Pro MiMo Max | 1050K | MiMo-V2.5 is a native omnimodal model by Xiaomi. It delivers Pro-level agentic performance at roughly half the inference cost, while surpassing MiMo-V… |
| Tencent: Hunyuan A13B Instruct tencent/hunyuan-a13b-instruct | — | 0.14/0.57 | 未上架 | — | — | 131K | Hunyuan-A13B is a 13B active parameter Mixture-of-Experts (MoE) language model developed by Tencent, with a total parameter count of 80B and support f… |
| Kwaipilot: KAT-Coder-Air V2.5 kwaipilot/kat-coder-air-v2.5 | — | 0.15/0.6 | 未上架 | — | — | 256K | KAT-Coder-Air V2.5 is a flagship-level Agentic Coding model that can directly hand over an entire issue or an entire business workflow to it, allowing… |
| Cohere: Command R (08-2024) cohere/command-r-08-2024 | — | 0.15/0.6 | 未上架 | — | — | 128K | command-r-08-2024 is an update of the [Command R](/models/cohere/command-r) with improved performance for multilingual retrieval-augmented generation … |
| Tencent: Hy3 preview tencent/hy3-preview | — | 0.18/0.6 | 0.172/0.572 | — | — | 262K | Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use.… |
| Meta: Llama Guard 4 12B meta-llama/llama-guard-4-12b | — | 0.18/0.18 | 未上架 | — | ZenMux Ultra | 1049K | Llama Guard 4 is a Llama 4 Scout-derived multimodal pretrained model, fine-tuned for content safety classification.… |
| StepFun: Step 3.7 Flash stepfun/step-3.7-flash | — | 0.2/1.15 | 0.2/1.15 | — | — | 262K | Step 3.7 Flash is StepFun's latest high-efficiency multimodal Mixture-of-Experts model.… |
| Meta: Llama 4 Maverick meta-llama/llama-4-maverick | — | 0.2/0.8 | 未上架 | — | ZenMux Ultra | 1049K | Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128… |
| MiniMax: MiniMax-01 minimax/minimax-01 | — | 0.2/1.1 | 未上架 | — | OpenCode Go ZenMux Ultra | 1000K | MiniMax-01 is a combines MiniMax-Text-01 for text generation and MiniMax-VL-01 for image understanding.… |
| MiniMax: MiniMax M2.5 minimax/minimax-m2.5 | — | 0.22/0.9 | 0.3/1.2 | — | OpenCode Go ZenMux Ultra | 205K | MiniMax-M2.5 is a SOTA large language model designed for real-world productivity.… |
| ByteDance Seed: Seed-2.0-Lite bytedance-seed/seed-2.0-lite | — | 0.25/2 | 未上架 | — | — | 262K | Seed-2.0-Lite is a versatile, cost‑efficient enterprise workhorse that delivers strong multimodal and agent capabilities while offering noticeably low… |
| ByteDance Seed: Seed 1.6 bytedance-seed/seed-1.6 | — | 0.25/2 | 未上架 | — | — | 262K | Seed 1.6 is a general-purpose model released by the ByteDance Seed team.… |
| DeepSeek: DeepSeek V3.1 deepseek/deepseek-chat-v3.1 | — | 0.25/0.95 | 0.28/1.11 | — | OpenCode Go ZenMux Ultra | 164K | DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates.… |
| MiniMax: MiniMax M2 minimax/minimax-m2 | — | 0.255/1.02 | 0.3/1.2 | — | OpenCode Go ZenMux Ultra | 205K | MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows.… |
| DeepSeek: DeepSeek V3 deepseek/deepseek-chat | — | 0.2574/1.0287 | 未上架 | — | OpenCode Go ZenMux Ultra | 164K | DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions.… |
| DeepSeek: DeepSeek V3 0324 deepseek/deepseek-chat-v3-0324 | — | 0.27/1.12 | 未上架 | — | OpenCode Go ZenMux Ultra | 164K | DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team.… |
| MiniMax: MiniMax M3 minimax/minimax-m3 | — | 0.3/1.2 | 0.27/1.08 | — | OpenCode Go ZenMux Ultra | 1049K | MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and i… |
| Kwaipilot: KAT-Coder-Pro V2 kwaipilot/kat-coder-pro-v2 | — | 0.3/1.2 | 未上架 | — | — | 262K | KAT-Coder-Pro V2 is the latest high-performance model in KwaiKAT’s KAT-Coder series, designed for complex enterprise-grade software engineering and Sa… |
| MiniMax: MiniMax M2.7 minimax/minimax-m2.7 | — | 0.3/1.2 | 0.3/1.2 | — | OpenCode Go ZenMux Ultra | 205K | MiniMax-M2.7 is a next-generation large language model designed for autonomous, real-world productivity and continuous improvement.… |
| MiniMax: MiniMax M2-her minimax/minimax-m2-her | — | 0.3/1.2 | 0.3/1.2 | — | OpenCode Go ZenMux Ultra | 66K | MiniMax M2-her is a dialogue-first large language model built for immersive roleplay, character-driven chat, and expressive multi-turn conversations.… |
| MiniMax: MiniMax M2.1 minimax/minimax-m2.1 | — | 0.3/1.2 | 0.3/1.2 | — | OpenCode Go ZenMux Ultra | 205K | MiniMax-M2.1 is a lightweight, state-of-the-art large language model optimized for coding, agentic workflows, and modern application development.… |
| Z.ai: GLM 4.6V z-ai/glm-4.6v | — | 0.3/0.9 | 0.1456/0.4367 | — | OpenCode Go ZenMux Ultra GLM Coding | 131K | GLM-4.6V is a large multimodal model designed for high-fidelity visual understanding and long-context reasoning across images, documents, and mixed me… |
| Amazon: Nova 2 Lite amazon/nova-2-lite-v1 | — | 0.3/2.5 | 未上架 | — | — | 1000K | Nova 2 Lite is a fast, cost-effective reasoning model for everyday workloads that can process text, images, and videos to generate text.… |
| Qwen2.5 72B Instruct qwen/qwen-2.5-72b-instruct | — | 0.36/0.4 | 未上架 | — | OpenCode Go ZenMux Ultra | 33K | Qwen2.5 72B is the latest series of Qwen large language models.… |
| Z.ai: GLM 4.7 z-ai/glm-4.7 | — | 0.4/1.75 | 0.2911/1.1645 | — | OpenCode Go ZenMux Ultra GLM Coding | 205K | GLM-4.7 is Z.ai’s latest flagship model, featuring upgrades in two key areas: enhanced programming capabilities and more stable multi-step reasoning/e… |
| Meta: Llama 3.1 70B Instruct meta-llama/llama-3.1-70b-instruct | — | 0.4/0.4 | 未上架 | — | ZenMux Ultra | 131K | Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors.… |
| Baidu: ERNIE 4.5 VL 424B A47B baidu/ernie-4.5-vl-424b-a47b | — | 0.42/1.25 | 未上架 | — | — | 123K | ERNIE-4.5-VL-424B-A47B is a multimodal Mixture-of-Experts (MoE) model from Baidu’s ERNIE 4.5 series, featuring 424B total parameters with 47B active p… |
| Xiaomi: MiMo-V2.5-Pro xiaomi/mimo-v2.5-pro | — | 0.435/0.87 | 0.44/0.88 | — | OpenCode Go MiMo Lite MiMo Standard MiMo Pro MiMo Max | 1050K | MiMo-V2.5-Pro is Xiaomi’s flagship model, delivering strong performance in general agentic capabilities, complex software engineering, and long-horizo… |
| Qwen: Qwen3.8 27B qwen/qwen3.8-27b | — | 0.45/3.2 | 未上架 | — | OpenCode Go ZenMux Ultra | 262K | Qwen3.8 27B is an open-weight dense vision-language model from Qwen.… |
| ByteDance Seed: Seed 2.1 Turbo bytedance-seed/seed-2-1-turbo | — | 0.5/2.5 | 未上架 | — | — | 262K | Seed 2.1 Turbo is a multimodal model from ByteDance Seed for coding and long-horizon agent workflows.… |
| ByteDance Seed: Seed-2.0-Code bytedance-seed/seed-2.0-code | — | 0.5/3 | 未上架 | — | — | 262K | Seed 2.0 Code is a model from ByteDance Seed optimized for agentic coding.… |
| Z.ai: GLM 5.2 z-ai/glm-5.2 | — | 0.5/3.15 | 0.98/3.08 | — | OpenCode Go ZenMux Ultra GLM Coding | 1049K | GLM 5.2 is a large-scale reasoning model from Z.ai. It supports text input and output with a 1M-token context window, and is suited for long-horizon a… |
| Z.ai: GLM 4.6 z-ai/glm-4.6 | — | 0.5/2 | 0.2911/1.1645 | — | OpenCode Go ZenMux Ultra GLM Coding | 205K | Compared with GLM-4.5, this generation brings several key improvements: Longer context window: The context window has been expanded from 128K to 200K … |
| DeepSeek: R1 0528 deepseek/deepseek-r1-0528 | — | 0.5/2.15 | 0.56/2.23 | — | OpenCode Go ZenMux Ultra | 164K | May 28th update to the [original DeepSeek R1](/deepseek/deepseek-r1) Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully … |
| MiniMax: MiniMax M1 minimax/minimax-m1 | — | 0.55/2.2 | 未上架 | — | OpenCode Go ZenMux Ultra | 1000K | MiniMax-M1 is a large-scale, open-weight reasoning model designed for extended context and high-efficiency inference.… |
| NVIDIA: Nemotron 3 Ultra nvidia/nemotron-3-ultra-550b-a55b | — | 0.6/3.6 | 未上架 | — | — | 512K | NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE).… |
| Z.ai: GLM 5 z-ai/glm-5 | — | 0.6/1.92 | 0.58/2.6 | — | OpenCode Go ZenMux Ultra GLM Coding | 205K | GLM-5 is Z.ai’s flagship open-source foundation model engineered for complex systems design and long-horizon agent workflows.… |
| MoonshotAI: Kimi K2 0905 moonshotai/kimi-k2-0905 | — | 0.6/2.5 | 未上架 | — | OpenCode Go ZenMux Ultra Kimi 行板 Kimi 中板 Kimi 小快板 Kimi 快板 | 262K | Kimi K2 0905 is the September update of [Kimi K2 0711](moonshotai/kimi-k2).… |
| Z.ai: GLM 4.5V z-ai/glm-4.5v | — | 0.6/1.8 | 未上架 | — | OpenCode Go ZenMux Ultra GLM Coding | 66K | GLM-4.5V is a vision-language foundation model for multimodal agent applications.… |
| Z.ai: GLM 4.5 z-ai/glm-4.5 | — | 0.6/2.2 | 0.2911/1.1645 | — | OpenCode Go ZenMux Ultra GLM Coding | 131K | GLM-4.5 is our latest flagship foundation model, purpose-built for agent-based applications.… |
| Qwen2.5 Coder 32B Instruct qwen/qwen-2.5-coder-32b-instruct | — | 0.66/1 | 未上架 | — | OpenCode Go ZenMux Ultra | 33K | Qwen2.5-Coder is the latest series of Code-Specific Qwen large language models (formerly known as CodeQwen).… |
| DeepSeek: R1 deepseek/deepseek-r1 | — | 0.7/2.5 | 未上架 | — | OpenCode Go ZenMux Ultra | 64K | DeepSeek R1 is here: Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens.… |
| Kwaipilot: KAT-Coder-Pro V2.5 kwaipilot/kat-coder-pro-v2.5 | — | 0.74/2.96 | 未上架 | — | — | 256K | KAT-Coder-Pro V2.5 is a flagship-level Agentic Coding model that can directly hand over an entire issue or an entire business workflow to it, allowing… |
| Qwen: Qwen2.5 VL 72B Instruct qwen/qwen2.5-vl-72b-instruct | — | 0.8/1 | 未上架 | — | OpenCode Go ZenMux Ultra | 128K | Qwen2.5-VL is proficient in recognizing common objects such as flowers, birds, fish, and insects.… |
| DeepSeek: R1 Distill Llama 70B deepseek/deepseek-r1-distill-llama-70b | — | 0.8/0.8 | 未上架 | — | OpenCode Go ZenMux Ultra | 8K | DeepSeek R1 Distill Llama 70B is a distilled large language model based on [Llama-3.3-70B-Instruct](/meta-llama/llama-3.3-70b-instruct), using outputs… |
| Amazon: Nova Pro 1.0 amazon/nova-pro-v1 | — | 0.8/3.2 | 未上架 | — | — | 300K | Amazon Nova Pro 1.0 is a capable multimodal model from Amazon focused on providing a combination of accuracy, speed, and cost for a wide range of task… |
| Z.ai: GLM 5.1 z-ai/glm-5.1 | — | 0.966/3.036 | 0.8781/3.5126 | — | OpenCode Go ZenMux Ultra GLM Coding | 205K | GLM-5.1 delivers a major leap in coding capability, with particularly significant gains in handling long-horizon tasks.… |
| Perplexity: Sonar perplexity/sonar | — | 1/1 | 未上架 | — | — | 127K | Sonar is lightweight, affordable, fast, and simple to use — now featuring citations and the ability to customize sources.… |
| Z.ai: GLM 5V Turbo z-ai/glm-5v-turbo | — | 1.2/4 | 0.726/3.1946 | — | OpenCode Go ZenMux Ultra GLM Coding | 203K | GLM-5V-Turbo is Z.ai’s first native multimodal agent foundation model, built for vision-based coding and agent-driven tasks.… |
| Z.ai: GLM 5 Turbo z-ai/glm-5-turbo | — | 1.2/4 | 0.73/3.19 | — | OpenCode Go ZenMux Ultra GLM Coding | 203K | GLM-5 Turbo is a new model from Z.ai designed for fast inference and strong performance in agent-driven environments such as OpenClaw scenarios.… |
| AI21: Jamba Large 1.7 ai21/jamba-large-1.7 | — | 2/8 | 未上架 | — | — | 256K | Jamba Large 1.7 is the latest model in the Jamba open family, offering improvements in grounding, instruction-following, and overall efficiency.… |
| Perplexity: Sonar Reasoning Pro perplexity/sonar-reasoning-pro | — | 2/8 | 未上架 | — | — | 128K | Note: Sonar Pro pricing includes Perplexity search pricing. See [details here](https://docs.perplexity.ai/guides/pricing#detailed-pricing-breakdown-fo… |
| Perplexity: Sonar Deep Research perplexity/sonar-deep-research | — | 2/8 | 未上架 | — | — | 128K | Sonar Deep Research is a research-focused model designed for multi-step retrieval, synthesis, and reasoning across complex topics.… |
| Amazon: Nova Premier 1.0 amazon/nova-premier-v1 | — | 2.5/12.5 | 未上架 | — | — | 1000K | Amazon Nova Premier is the most capable of Amazon’s multimodal models for complex reasoning tasks and for use as the best teacher for distilling custo… |
| Cohere: Command A cohere/command-a | — | 2.5/10 | 未上架 | — | — | 256K | Command A is an open-weights 111B parameter model with a 256k context window focused on delivering great performance across agentic, multilingual, and… |
| Cohere: Command R+ (08-2024) cohere/command-r-plus-08-2024 | — | 2.5/10 | 未上架 | — | — | 128K | command-r-plus-08-2024 is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as c… |
| Perplexity: Sonar Pro Search perplexity/sonar-pro-search | — | 3/15 | 未上架 | — | — | 200K | Exclusively available on the OpenRouter API, Sonar Pro's new Pro Search mode is Perplexity's most advanced agentic search system.… |
| Perplexity: Sonar Pro perplexity/sonar-pro | — | 3/15 | 未上架 | — | — | 200K | Note: Sonar Pro pricing includes Perplexity search pricing. See [details here](https://docs.perplexity.ai/guides/pricing#detailed-pricing-breakdown-fo… |
| Google: Lyria 3 Pro Preview google/lyria-3-pro-preview | — | / | 未上架 | — | ZenMux Ultra | 1049K | Full-length songs are priced at $0.08 per song. Lyria 3 is Google's family of music generation models, available through the Gemini API.… |
| Google: Lyria 3 Clip Preview google/lyria-3-clip-preview | — | / | 未上架 | — | ZenMux Ultra | 1049K | 30 second duration clips are priced at $0.04 per clip. Lyria 3 is Google's family of music generation models, available through the Gemini API.… |
| Z.AI: GLM 5.3 z-ai/glm-5.3 | — | — | 1.4/4.4 | — | OpenCode Go ZenMux Ultra GLM Coding | 1000K | |
| inclusionAI: Ling-3.0-tiny inclusionai/ling-3.0-tiny | — | — | / | — | — | 262K | |
| MoonshotAI: Kimi K2.7 Code HighSpeed moonshotai/kimi-k2.7-code-highspeed | — | — | 1.9/8 | — | OpenCode Go ZenMux Ultra Kimi 行板 Kimi 中板 Kimi 小快板 Kimi 快板 | 262K | |
| Baidu: ERNIE 5.1 baidu/ernie-5.1 | — | — | 0.588/2.646 | — | — | 128K | |
| MiniMax: MiniMax M2.7 highspeed minimax/minimax-m2.7-highspeed | — | — | 0.611/2.4439 | — | OpenCode Go ZenMux Ultra | 205K | |
| inclusionAI: LLaDA2.1-flash inclusionai/llada2.1-flash | — | — | 0.28/2.85 | — | — | 32K | |
| xAI: Grok 4.2 Fast x-ai/grok-4.2-fast | — | — | 2/6 | — | ZenMux Ultra | 2000K | |
| xAI: Grok 4.2 Fast Non Reasoning x-ai/grok-4.2-fast-non-reasoning | — | — | 2/6 | — | ZenMux Ultra | 2000K | |
| Qwen: Qwen3.5-Flash qwen/qwen3.5-flash | — | — | 0.1/0.4 | — | OpenCode Go ZenMux Ultra | 1024K | |
| Qwen: Qwen3.5-Plus qwen/qwen3.5-plus | — | — | 0.4/2.4 | — | OpenCode Go ZenMux Ultra | 1000K | |
| MiniMax: MiniMax M2.5 highspeed minimax/minimax-m2.5-lightning | — | — | / | — | OpenCode Go ZenMux Ultra | 205K | |
| Baidu: ERNIE 5.0 baidu/ernie-5.0-thinking-preview | — | — | 0.84/3.37 | — | — | 128K | |
| Z.AI: GLM 4.7 Flash z-ai/glm-4.7-flash-free | — | — | / | — | OpenCode Go ZenMux Ultra GLM Coding | 200K | |
| Z.AI: GLM 4.7 FlashX z-ai/glm-4.7-flashx | — | — | 0.0728/0.4367 | — | OpenCode Go ZenMux Ultra GLM Coding | 200K | |
| Qwen: Qwen3 VL Embedding qwen/qwen3-vl-embedding | — | — | 0.1/ | — | OpenCode Go ZenMux Ultra | 8K | |
| Z.AI: GLM 4.6V Flash z-ai/glm-4.6v-flash-free | — | — | / | — | OpenCode Go ZenMux Ultra GLM Coding | 200K | |
| Z.AI: GLM 4.6V FlashX z-ai/glm-4.6v-flash | — | — | 0.0218/0.2184 | — | OpenCode Go ZenMux Ultra GLM Coding | 200K | |
| Qwen: Qwen3-VL-Plus qwen/qwen3-vl-plus | — | — | 0.2/1.6 | — | OpenCode Go ZenMux Ultra | 262K | |
| Baidu: ERNIE-X1.1-Preview baidu/ernie-x1.1-preview | — | — | 0.14/0.56 | — | — | 66K | |
| Qwen: Qwen3 ASR Flash qwen/qwen3-asr-flash | — | — | / | — | OpenCode Go ZenMux Ultra | 1000K |
单位：美元 / 每百万 token，标准档（不含 batch、缓存命中、快速模式）。 「官方」为人工维护的标价，是优惠标签的判定基准；OpenRouter 一列在你打开页面时实时刷新。 点模型名可跳到它在 OpenRouter 模型库 的详情页，那里有分 provider 的价格、延迟与吞吐。 数据源 OpenRouter · ZenMux。 价格随时可能变动，下单前请以各平台页面为准。
订阅套餐
订阅和 API 是两种不同的东西，所以是两张表：订阅按「美元/月」卖一个服务（含网页端、App、各种额外功能），API 按「美元/每百万 token」卖调用量。塞进同一张表的同一列只会误导。
两者唯一能挂钩的地方是「等值 API 用量」——把月费按主力模型的输入单价折算成 token 量，给你一个数量级参照。注意那是纯输入的量级，真实用量是输入输出混合的，而输出通常贵 5 倍。
| 套餐 | 月费原生币种 | 额度 | 包含 | 来源 | 等值 API 用量仅美元定价可算 |
|---|---|---|---|---|---|
| Anthropic Claude Pro 核对于 2026-08-18 | $20/月 | 免费版约 2 倍用量；5 小时窗口 + 周上限 | 网页端 / 桌面端 / App / Claude Code年付 $17/月。可用 Fable、Opus、Sonnet、Haiku 全系 | claude.com/pricing | ≈ 4.0M tokens 按 anthropic/claude-opus-5 输入价 $5 |
| Anthropic Claude Max 5x 核对于 2026-08-18 | $100/月 | Pro 的 5 倍用量 | 同 Pro，高峰期优先官网标「From $100」。另有 Max 20x 档，官网未直接列月费，需在购买页确认 | claude.com/pricing | ≈ 20.0M tokens 按 anthropic/claude-opus-5 输入价 $5 |
| OpenAI ChatGPT Go 核对于 2026-08-18 | $8/月 | 低价入门档，额度低于 Plus | 网页端 / App美国区 $8/月，其它区域定价不同。covers 留空：官方未明确列出可用模型范围 | openai.com/index/introducing-chatgpt-go | ≈ 4.6M tokens 按 openai/gpt-5.2 输入价 $1.75 |
| OpenAI ChatGPT Plus 核对于 2026-08-18 | $20/月 | 扩展额度，含 GPT-5.2 Thinking | 网页端 / App / Codex 编码 Agent / 可选旧版模型 | help.openai.com/en/articles/6950777 | ≈ 11.4M tokens 按 openai/gpt-5.2 输入价 $1.75 |
| OpenAI ChatGPT Pro 核对于 2026-08-18 | $200/月 | 最高额度，含 GPT-5.2 Pro | 最大记忆与上下文 / 新功能抢先另有 $100 档，额度低于 $200 档 | help.openai.com/en/articles/9793128 | ≈ 9.5M tokens 按 openai/gpt-5.2-pro 输入价 $21 |
| OpenCode Go 核对于 2026-08-18 | $10/月 | 官方称按额度约给到月费的 6 倍价值 | OpenAI 兼容端点，可接 OpenClaw / Hermes / Pi 等任意 Agent首月 $5，之后 $10/月。含 Kimi K3、Grok 4.5、Qwen3.8 Max、GLM-5.2、DeepSeek V4 Pro、MiniMax M3、GPT 5.6 Luna、MiMo-V2.5 等开放模型 | opencode.ai/go | ≈ 71.4M tokens 按 deepseek/deepseek-v4-flash 输入价 $0.14 |
| ZenMux Starter 邀 核对于 2026-08-18 | $20/月 | 50 Flows/月，每月 4 次窗口重置 | 基础模型 + 限时开放的旗舰模型covers 留空：官方只写「基础+限时旗舰」，无确切名单且随时调整 | zenmux.ai/docs/zh/guide/subscription.html | ≈ 4.0M tokens 按 anthropic/claude-opus-5 输入价 $5 |
| ZenMux Max 邀 核对于 2026-08-18 | $100/月 | 300 Flows/月，每月 3 次窗口重置 | 基础模型 + 高级模型（覆盖多数主流旗舰）同上，具体名单以官网套餐卡片为准 | zenmux.ai/docs/zh/guide/subscription.html | ≈ 20.0M tokens 按 anthropic/claude-opus-5 输入价 $5 |
| ZenMux Ultra 邀 核对于 2026-08-18 | $200/月 | 800 Flows/月，每月 2 次窗口重置 | 全量模型官方写「全量模型」，所以列了全部收录厂商。1 Flow ≈ $0.03283（官方换算，会调整） | zenmux.ai/docs/zh/guide/subscription.html | ≈ 40.0M tokens 按 anthropic/claude-opus-5 输入价 $5 |
| 小米 MiMo Token Plan Lite 核对于 2026-08-18 | ¥39/月 ≈ $6 | 4.1B Credits/月 | 可接 OpenCode / OpenClaw / Claude Code 等主流工具国际开发者 $6/月。夜间 0.8 折、首购 88 折、连续包年 88 折 | mimo.mi.com/docs/zh-CN/tokenplan | 仅人民币定价 |
| 小米 MiMo Token Plan Standard 核对于 2026-08-18 | ¥99/月 ≈ $16 | 11B Credits/月 | 同 Lite，额度 2.7 倍 | mimo.mi.com/docs/zh-CN/tokenplan | 仅人民币定价 |
| 小米 MiMo Token Plan Pro 核对于 2026-08-18 | ¥329/月 ≈ $50 | 38B Credits/月 | 深度嵌入工作流的专业用户档 | mimo.mi.com/docs/zh-CN/tokenplan | 仅人民币定价 |
| 小米 MiMo Token Plan Max 核对于 2026-08-18 | ¥659/月 ≈ $100 | 82B Credits/月 | 全天高强度使用档六款模型全档位可用：mimo-v2.5-pro / v2.5 / asr / tts 系列三款 | mimo.mi.com/docs/zh-CN/tokenplan | 仅人民币定价 |
| 月之暗面 Kimi Andante 行板 核对于 2026-08-18 | ¥49/月 | 约 30 个 Agent 用量；Kimi Code 1 倍额度 | 4 倍速优先队列Kimi Code 另有 5 小时/周限额，独立于共享额度池。年付更优惠 | kimi.com/zh-cn/help/membership/membership-pricing | 仅人民币定价 |
| 月之暗面 Kimi Moderato 中板 核对于 2026-08-18 | ¥99/月 | 约 60 个 Agent 用量；Kimi Code 4 倍额度 | 2 个并行任务 | kimi.com/zh-cn/help/membership/membership-pricing | 仅人民币定价 |
| 月之暗面 Kimi Allegretto 小快板 核对于 2026-08-18 | ¥199/月 | 约 150 个 Agent 用量；Kimi Code 20 倍额度 | 含 Kimi Claw 部署 | kimi.com/zh-cn/help/membership/membership-pricing | 仅人民币定价 |
| 月之暗面 Kimi Allegro 快板 核对于 2026-08-18 | ¥699/月 | 约 360 个 Agent 用量；Kimi Code 60 倍额度 | 4 个并行任务年付最高立省 ¥1,680 | kimi.com/zh-cn/help/membership/membership-pricing | 仅人民币定价 |
| 智谱 GLM Coding Plan 核对于 2026-08-18 | ¥118/月 | 分 Lite / Pro / Max 三档：5 小时积分 2,000 / 12,000 / 28,000，周积分 10,000 / 60,000 / 140,000 | 支持 GLM-5.3、GLM-5-Turbo、GLM-4.7；透明积分制，MCP 能力也计入⚠️ ¥118 是「起步价」（来自 IT之家 报道）。官方文档只公布了三档的积分额度、未公布分档价格 —— 分档月费需去官网定价页确认。包年/包季曾有 7 折/8 折限时优惠 | docs.bigmodel.cn/cn/coding-plan/overview + ithome.com/0/983/934.htm | 仅人民币定价 |
「等值 API 用量」= 月费 ÷ 该模型 API 输入单价，只算纯输入， 是给你一个数量级参照，不是真实可用量 —— 实际用量是输入输出混合的，输出通常贵 5 倍。 订阅价格全部人工维护，没有自动更新；各家改价不会有任何提示，以官方页面为准。
模型任务成功率判断
模型的使用成本除了价格之外，也应该关注它的成功率。便宜模型成功率低，导致多轮次修改返工，综合成本反而更高；Agent的使用过程这个成本会更加被放大。
推荐下面网站查询选中模型的执行成功率，建议成功率不要低于60%。
【成功率查询】：www.swebench.com/index.html
几个容易踩的坑
- 同一模型在同一平台常有多档价。batch 通常是标准价五折，缓存命中更低，而快速模式反而更贵——比如 Claude Opus 5 的 fast 模式是 $10/$50，标准档只要 $5/$25。这张表只列标准档。
- 聚合平台会滞后，也会留着已下线的条目。实测发现过 OpenRouter 仍在列
claude-opus-4.7-fast，但该模型的快速模式已经下线、调用会直接报错。写文章引用具体数字前，建议点进对应平台确认一次。 - 上下文长度是模型的理论最大值，实际可用可能受平台配置限制。