LimitWatch

AI model prices compared

Per-million-token input and output list prices (USD) for models we track, grouped by model with the Direct provider first, then Aggregator routes. Direct prices come from each provider's official pricing page (stored snapshot); aggregator prices from OpenRouter, Hugging Face and Vercel AI Gateway's official model lists, refreshed at each build. Every row shows its source and the date last verified (US Central). For the percentage difference between an aggregator and buying direct, see aggregators vs direct (difference = (route − direct) ÷ direct).

Model · providerInput $/1MOutput $/1MRouteVerifiedSource and history
gpt-6-lunaOpenAI (ChatGPT)$0.10$0.50Direct2026-10-22
Llama 3.3 70B InstructOpenRouter$0.10$0.32Aggregator2026-10-23
Llama 3.3 70B InstructHugging Face (novita)$0.14$0.40Aggregator2026-10-23
Llama 3.3 70B InstructVercel AI Gateway$0.72$0.72Aggregator2026-10-23
Llama 3.3 70B InstructHugging Face (ovhcloud)$0.74$0.74Aggregator2026-10-23
Llama 3.3 70B InstructHugging Face (scaleway)$1.03$1.03Aggregator2026-10-23
deepseek-flashDeepSeekoff-peak $0.15/$0.60$0.30$1.20Direct2026-10-22
deepseek-v4.1-flashOllama$0.30$1.20Direct2026-10-22
Gemini 2.5 FlashGoogle (Gemini)text/image/video input; free tier: free of charge$0.30$2.50Direct2026-10-22
Gemini 2.5 FlashOpenRouter$0.30$2.50Aggregator2026-10-23
Gemini 2.5 FlashVercel AI Gateway$0.30$2.50Aggregator2026-10-23
Mistral LargeMistral AI$0.50$1.50Direct2026-10-22
DeepSeek v3.2 (US regions, standard)Amazon Bedrock (AWS)$0.62$1.85Direct2026-10-22
deepseek-v4-proOllama$1.32$3.96Direct2026-10-22
Claude Sonnet 5OpenRouter$2.00$10.00Aggregator2026-10-23
Claude Sonnet 5Vercel AI Gateway$2.00$10.00Aggregator2026-10-23
Claude Sonnet 5.5Anthropic (Claude)$2.00$10.00Direct2026-10-22
Claude Sonnet 5.5OpenRouter$2.00$10.00Aggregator2026-10-23
Claude Sonnet 5.5Vercel AI Gateway$2.00$10.00Aggregator2026-10-23
Gemini 4 ArgonGoogle (Gemini)through Nov 30, 2026; $4.00 / $20.00 from Dec 1, 2026$2.00$10.00Direct2026-10-22
gpt-6.1-solOpenAI (ChatGPT)$2.00$10.00Direct2026-10-22
Grok 4.7xAI (Grok)$2.00$6.00Direct2026-10-22
Qwen3.8-2.4T-A95BTogether AIoutput price split across lines in snapshot; not verified$2.00n/a†Direct2026-10-22
GPT-4oOpenRouter$2.50$10.00Aggregator2026-10-23
GPT-4oVercel AI Gateway$2.50$10.00Aggregator2026-10-23
Kimi K3Together AI$3.00$15.00Direct2026-10-22
Claude Opus 5.5Anthropic (Claude)$4.00$20.00Direct2026-10-22
Claude Fable 5.1Anthropic (Claude)$10.00$50.00Direct2026-10-22
gpt-6-astraOpenAI (ChatGPT)$10.00$50.00Direct2026-10-22

† n/a: not stated by the source, not applicable to this row, or the page could not be fetched; the reason is shown on the row where one applies.

Not listed (no verifiable per-token price): Azure OpenAI (Microsoft), Cohere, Cursor (Anysphere), Fireworks AI, GitHub Copilot, Groq, Hugging Face, Meta (Llama), OpenRouter, Perplexity, Poe (Quora), Replit, T3 Chat (T3 Tools), v0 (Vercel), Vercel AI Gateway, Windsurf. See each company page for plans and limits.