A clean, neutral benchmark of foundation model token costs, prompt caching rates, and context limits across 34 foundation models. Normalized per 1 Million tokens.
| Model | Provider | Context | Input / 1M | Output / 1M | Cache Read / 1M | Cache Write / 1M |
|---|---|---|---|---|---|---|
| DeepSeek V4.1 Flash | DeepSeek | 1049k | 0.1500 | 0.6000 | $0.0030 | — |
| Claude Opus 5.5Reasoning | Anthropic | 1000k | 4.0000 | 20.0000 | $0.2000 | $5.0000 |
| GLM 5.3 FlashReasoning | Zhipu AI | 1311k | 0.1500 | 0.5000 | $0.0300 | $0.0750 |
| Kimi K3.1 ProReasoning | Moonshot | 1049k | 2.0000 | 10.0000 | $0.2000 | — |
| MiniMax M3.1 FlashReasoning | MiniMax | 1049k | 0.1500 | 0.6000 | $0.0150 | — |
| Mercury 3Reasoning | Inception | 1000k | 0.1000 | 0.3500 | $0.0100 | — |
| Step 5 PreviewReasoning | StepFun | 1000k | 1.0000 | 2.7000 | $0.1000 | — |
| MiMo-V2.6 Flash | Xiaomi | 1050k | 0.1400 | 0.2800 | $0.0028 | — |
| GPT-6 AstraReasoning | OpenAI | 1050k | 10.0000 | 50.0000 | $1.0000 | $12.5000 |
| Claude Fable 5.1Reasoning | Anthropic | 1000k | 10.0000 | 50.0000 | $0.2500 | $12.5000 |
| Claude Sonnet 5.5Reasoning | Anthropic | 1000k | 2.0000 | 10.0000 | $0.1000 | $2.5000 |
| GPT-6 Luna | OpenAI | 1050k | 0.1000 | 0.5000 | $0.0100 | $0.1250 |
| Qwen 3.8 27BReasoning | Qwen | 1000k | 0.2000 | 2.5000 | $0.0900 | — |
| GPT-6.1 SolReasoning | OpenAI | 1050k | 2.0000 | 10.0000 | $0.1000 | $2.5000 |
| Qwen 3.8 Flash | Qwen | 1000k | 0.1500 | 0.4700 | $0.0160 | — |
| Nemotron 3 UltraReasoning | NVIDIA | 256k | 0.6000 | 2.4000 | $0.1200 | — |
| Gemini 4 ArgonReasoning | 1049k | 2.0000 | 10.0000 | $0.1000 | $0.1875 | |
| MiMo-V2.6 ProReasoning | Xiaomi | 1050k | 0.4350 | 0.8700 | $0.0036 | — |
| GLM 5.3Reasoning | Zhipu AI | 1311k | 0.0699 | 1.4000 | $0.0689 | $0.8775 |
| DeepSeek V4 ProReasoning | DeepSeek | 1049k | 0.9550 | 1.9110 | $0.0796 | $0.9480 |
| Llama 4 Scout | Meta | 1300k | 0.1000 | 0.3000 | — | — |
| Gemini 3.8 FlashReasoning | 1049k | 0.8250 | 4.1250 | $0.0825 | $0.0417 | |
| Grok 4.7Reasoning | xAI | 500k | 2.0000 | 6.0000 | $0.5000 | $2.0000 |
| GLM 5.3 PrimeReasoning | Zhipu AI | 1000k | 2.8000 | 8.8000 | $0.5600 | $2.8000 |
| Qwen 3.8 MaxReasoning | Qwen | 1000k | 2.0000 | 6.0000 | $0.2500 | — |
| Claude Sonnet 4.6Reasoning | Anthropic | 1000k | 3.0000 | 15.0000 | $0.3000 | $3.7500 |
| Gemini 3.5 Flash-Lite | 1000k | 0.3000 | 2.5000 | $0.0300 | $0.0833 | |
| Ling 3.0 Flash | InclusionAI | 262k | 0.0210 | 0.0630 | $0.0042 | — |
| Llama 4 Maverick | Meta | 1000k | 0.1875 | 0.6525 | $0.0500 | — |
| Claude Haiku 5.5Reasoning | Anthropic | 1000k | 0.1000 | 0.5000 | $0.0100 | $0.1250 |
| Mistral Large 4Reasoning | Mistral | 1049k | 1.3600 | 4.1800 | $0.1400 | — |
| Nemotron 3.5 Super VLReasoning | NVIDIA | 1049k | 0.1000 | 0.4000 | $0.0800 | — |
| Muse WonderReasoning | Meta | 1049k | 4.0000 | 20.0000 | $0.2000 | — |
| Ling 3.1 Flash | InclusionAI | 262k | 0.0750 | 0.2250 | $0.0150 | — |
Download the entire real-time model catalog as structured JSON (34 models).
Download models.json →Machine-readable index optimized for AI answer engines and crawlers.
View /llms.txt →Complete search engine index covering all programmatic model routes.
View /sitemap.xml →