Free · updated January 2026
LLM pricing catalog
19 models from 10 providers: current $/1M input and output pricing, tracked price history, context windows and benchmark-informed quality scores per task. Click any model for its full page, or head to the cost estimator to price a real prompt against all of them.
Score for
| Trend | Speed | Privacy | Notes | ||||||
|---|---|---|---|---|---|---|---|---|---|
GPT-5 nanoOpenAI | $0.05 | $0.40 | $0.14 | 400k | 72 | Blazing | Standard | ultra-cheaphigh volume | |
Mistral Small 3.2Mistral | $0.10 | $0.30 | $0.15 | 128k | 74 | Blazing | Self-host | openopen weightscheap | |
Gemini 2.5 Flash-LiteGoogle | $0.10 | $0.40 | $0.18 | 1M | 70 | Blazing | Standard | ultra-cheaphigh volume | |
| $0.20 | $0.50 | $0.28 | 2M | 82 | Fast | Standard | 2M contextcheap | ||
DeepSeek V3.2best valueDeepSeek | $0.28 | $0.42 | $0.32 | 128k | 89 | Brisk | Self-host | openopen weightsbest value reasoning | |
Qwen3 235BAlibaba (Qwen) | $0.22 | $0.88 | $0.39 | 256k | 86 | Brisk | Self-host | openopen weightsmultilingual | |
Llama 4 MaverickMeta | $0.27 | $0.85 | $0.42 | 1M | 80 | Fast | Self-host | openopen weights1M context | |
Llama 3.3 70BMeta | $0.54 | $0.68 | $0.58 | 128k | 76 | Blazing | Self-host | openopen weightsfast on Groq | |
GPT-5 miniOpenAI | $0.25 | $2.00 | $0.69 | 400k | 85 | Fast | Standard | workhorsecheap | |
Mistral Medium 3Mistral | $0.40 | $2.00 | $0.80 | 128k | 83 | Fast | Strong | EU hostingstrong translation | |
Gemini 2.5 FlashGoogle | $0.30 | $2.50 | $0.85 | 1M | 84 | Blazing | Standard | fast1M context | |
GLM-4.6Zhipu (GLM) | $0.60 | $2.20 | $1.00 | 200k | 88 | Brisk | Self-host | openopen weightsstrong coding value | |
Kimi K2Moonshot | $0.60 | $2.50 | $1.07 | 256k | 87 | Brisk | Self-host | openopen weightsagents | |
Claude Haiku 4.5Anthropic | $1.00 | $5.00 | $2.00 | 200k | 88 | Blazing | Strong | fastnear-frontier value | |
GPT-5.1OpenAI | $1.25 | $10.0 | $3.44 | 400k | 94 | Brisk | Standard | frontierreasoning | |
Gemini 3 ProGoogle | $2.00 | $12.0 | $4.50 | 1M | 95 | Brisk | Standard | frontierbest vision | |
Claude Sonnet 4.5Anthropic | $3.00 | $15.0 | $6.00 | 200k | 95 | Brisk | Strong | balanced flagshipagents | |
Grok 4.1xAI | $3.00 | $15.0 | $6.00 | 256k | 92 | Brisk | Standard | frontierreasoning | |
Claude Opus 4.5Anthropic | $5.00 | $25.0 | $10.0 | 200k | 99 | Measured | Strong | frontierbest coding |
19 of 19 models shown. Prices are list $/1M tokens; "typical" weights input 3:1 over output. The trend sparkline plots blended price since we started tracking the catalog; hover it for the last change date. Task scores are benchmark-informed 0-100 estimates, so validate on your own evals before migrating traffic.