# LLMPrices
> LLMPrices tracks API prices for 461 large language models from 66 providers, normalized to USD per 1M tokens (input, output and cached input). Data is refreshed daily from OpenRouter's public catalog; this snapshot is dated October 21, 2026. The cheapest paid model is IBM: Granite 4.0 Micro at $0.017 input per 1M tokens; 20 models are free.

## Key pages
- [All LLM API prices](https://llmprices.net/): sortable table of all 461 models with input, output, cached-input prices and context windows
- [Cheapest LLM APIs](https://llmprices.net/cheapest.html): rankings of the 25 cheapest models by input, output and blended price
- [Official prices](https://llmprices.net/official-prices.html): 78 flagship models whose prices were verified against each provider's own pricing page (72 match the catalog, 6 differ)
- [LLM API cost calculator](https://llmprices.net/calculator.html): estimate monthly cost for any model from token volumes
- [Free LLM API models](https://llmprices.net/free-models.html): all 20 models priced at $0
- [Largest context windows](https://llmprices.net/largest-context.html): models ranked by context size, led by SpaceXAI: Grok 4.20 Multi-Agent at 2M
- [Pricing FAQ](https://llmprices.net/faq.html): how per-token billing, caching and blended prices work
- [Methodology](https://llmprices.net/methodology.html): data sources, normalization rules and limitations
- [Full price table in Markdown](https://llmprices.net/llms-full.txt): the complete current price table in one file

## Top providers
- [openai API pricing](https://llmprices.net/providers/openai.html): 104 models from openai, cheapest OpenAI: gpt-oss-20b at $0.018 input per 1M tokens
- [qwen API pricing](https://llmprices.net/providers/qwen.html): 43 models from qwen, cheapest Qwen: Qwen3.7 Flash at $0.03 input per 1M tokens
- [google API pricing](https://llmprices.net/providers/google.html): 42 models from google, cheapest Google: Gemma 4 26B A4B  (free) at $0 (free) input per 1M tokens
- [anthropic API pricing](https://llmprices.net/providers/anthropic.html): 30 models from anthropic, cheapest Anthropic: Claude Haiku 5.5 (batch) at $0.05 input per 1M tokens
- [mistralai API pricing](https://llmprices.net/providers/mistralai.html): 26 models from mistralai, cheapest Mistral: Mistral Nemo at $0.029 input per 1M tokens
- [z-ai API pricing](https://llmprices.net/providers/z-ai.html): 18 models from z-ai, cheapest Z.ai: GLM 5.2 at $0.0341 input per 1M tokens
- [deepseek API pricing](https://llmprices.net/providers/deepseek.html): 15 models from deepseek, cheapest DeepSeek: DeepSeek V4.1 Flash at $0.02394 input per 1M tokens
- [nvidia API pricing](https://llmprices.net/providers/nvidia.html): 12 models from nvidia, cheapest NVIDIA: Nemotron 3.5 Super VL (free) at $0 (free) input per 1M tokens
- [minimax API pricing](https://llmprices.net/providers/minimax.html): 9 models from minimax, cheapest MiniMax: MiniMax M3.1 Flash (free) at $0 (free) input per 1M tokens
- [meta-llama API pricing](https://llmprices.net/providers/meta-llama.html): 8 models from meta-llama, cheapest Meta: Llama 3.2 1B Instruct at $0.027 input per 1M tokens

## Cheapest models right now
- [MiniMax: MiniMax M3.1 Flash (free) pricing](https://llmprices.net/models/minimax/minimax-m3.1-flash:free.html): $0 (free) input / $0 (free) output per 1M tokens, 1,000,000-token context
- [XingChen: Xing4.0 29B A4B (free) pricing](https://llmprices.net/models/xingchen-agi/xing4.0-29b-a4b:free.html): $0 (free) input / $0 (free) output per 1M tokens, 262,144-token context
- [NVIDIA: Nemotron 3.5 Super VL (free) pricing](https://llmprices.net/models/nvidia/nemotron-3.5-super-vl:free.html): $0 (free) input / $0 (free) output per 1M tokens, 1,000,000-token context
- [Dots Studio: Dots3-Note Preview (free) pricing](https://llmprices.net/models/dots-studio/dots-3-note-preview:free.html): $0 (free) input / $0 (free) output per 1M tokens, 512,000-token context
- [LiquidAI: LFM2.5-2.6B (free) pricing](https://llmprices.net/models/liquid/lfm-2.5-2.6b:free.html): $0 (free) input / $0 (free) output per 1M tokens, 65,536-token context
- [NVIDIA: Nemotron 3.5 Lightning (free) pricing](https://llmprices.net/models/nvidia/nemotron-3.5-lightning:free.html): $0 (free) input / $0 (free) output per 1M tokens, 1,000,000-token context
- [Thinking Machines: Inkling Small (free) pricing](https://llmprices.net/models/thinkingmachines/inkling-small:free.html): $0 (free) input / $0 (free) output per 1M tokens, 1,048,576-token context
- [Poolside: Laguna S 2.1 (free) pricing](https://llmprices.net/models/poolside/laguna-s-2.1:free.html): $0 (free) input / $0 (free) output per 1M tokens, 262,144-token context
- [Thinking Machines: Inkling (free) pricing](https://llmprices.net/models/thinkingmachines/inkling:free.html): $0 (free) input / $0 (free) output per 1M tokens, 1,048,576-token context
- [Poolside: Laguna XS 2.1 (free) pricing](https://llmprices.net/models/poolside/laguna-xs-2.1:free.html): $0 (free) input / $0 (free) output per 1M tokens, 262,144-token context