AI model pricing changelog
Daily price changes across 25+ inference providers. Generated from snapshot diffs above a 5% threshold — subscribe to the RSS feed.
May 27, 2026
DeepInfra added DeepSeek V3 at $0.32/M input, $0.89/M output.
DeepInfra added Llama 3.1 70B Instruct at $0.40/M input, $0.40/M output.
DeepInfra added DeepSeek R1 Distill Llama 70B at $0.70/M input, $0.80/M output.
DeepInfra added Llama 3.1 8B Instruct at $0.0200/M input, $0.0500/M output.
DeepInfra added Qwen 2.5 72B Instruct at $0.36/M input, $0.40/M output.
May 18, 2026
Hyperbolic added Mixtral 8x7B Instruct at $0.50/M input, $0.50/M output.
Hyperbolic added Llama 3.1 405B Instruct at $4.00/M input, $4.00/M output.
Hyperbolic added Qwen 2.5 72B Instruct at $0.40/M input, $0.40/M output.
Cerebras Inference added Qwen 3 32B Instruct at $0.40/M input, $0.80/M output.
Hyperbolic added Llama 3.1 8B Instruct at $0.10/M input, $0.10/M output.
Hyperbolic added Mixtral 8x22B Instruct at $0.60/M input, $0.60/M output.
Cerebras Inference added Llama 3.1 70B Instruct at $0.60/M input, $0.60/M output.
Hyperbolic added DeepSeek V3 at $0.25/M input, $0.25/M output.
Hyperbolic added DeepSeek R1 Distill Llama 70B at $0.20/M input, $0.20/M output.
Cerebras Inference added Llama 3.3 70B Instruct at $0.85/M input, $1.20/M output.
Hyperbolic added Llama 3.3 70B Instruct at $0.40/M input, $0.40/M output.
Hyperbolic added DeepSeek R1 at $2.00/M input, $2.00/M output.
Hyperbolic added Qwen 3 72B Instruct at $0.40/M input, $0.40/M output.
Cerebras Inference added Mistral Large 2.
Hyperbolic added Llama 3.1 70B Instruct at $0.40/M input, $0.40/M output.
Cerebras Inference added Llama 3.1 8B Instruct at $0.10/M input, $0.10/M output.
May 17, 2026
Together AI added Llama 3.1 405B Instruct at $3.50/M input, $3.50/M output.
Together AI added Mixtral 8x22B Instruct at $1.20/M input, $1.20/M output.
Together AI added DeepSeek V3.2 at $0.27/M input, $1.10/M output.
Together AI added Llama 3.3 70B Instruct at $0.88/M input, $0.88/M output.
Together AI added Qwen 3 72B Instruct at $0.29/M input, $0.29/M output.
May 16, 2026
Groq added Qwen 3 32B Instruct at $0.29/M input, $0.59/M output.
Groq added Llama 3.3 70B Instruct at $0.59/M input, $0.79/M output.
Groq added Mixtral 8x7B Instruct at $0.24/M input, $0.24/M output.
Groq added Llama 3.1 70B Instruct at $0.59/M input, $0.79/M output.
Groq added DeepSeek R1 Distill Llama 70B at $0.75/M input, $0.99/M output.
Groq added Gemma 2 9B IT at $0.0800/M input, $0.0800/M output.
Groq added Llama 3.1 8B Instruct at $0.0500/M input, $0.0800/M output.
OpenRouter added DeepSeek R1 at $0.50/M input, $2.30/M output.
OpenRouter added DeepSeek V3 at $0.27/M input, $1.10/M output.
OpenRouter added Llama 3.1 8B Instruct at $0.0600/M input, $0.0900/M output.
OpenRouter added Command R+ at $2.50/M input, $10.00/M output.
OpenRouter added Llama 3.1 405B Instruct at $3.50/M input, $10.50/M output.
OpenRouter added Llama 3.3 70B Instruct at $0.27/M input, $0.85/M output.
OpenRouter added Qwen 3 72B Instruct at $0.27/M input, $0.85/M output.
OpenRouter added Mistral Small 3 at $0.10/M input, $0.30/M output.
OpenRouter added Mixtral 8x22B Instruct at $0.65/M input, $0.65/M output.
OpenRouter added Qwen 2.5 72B Instruct at $0.22/M input, $0.75/M output.
OpenRouter added Mistral Large 2 at $1.80/M input, $5.40/M output.
OpenRouter added Qwen 3 32B Instruct at $0.14/M input, $0.55/M output.
DeepInfra added DeepSeek R1 Distill Llama 70B at $0.28/M input, $0.55/M output.
DeepInfra added Llama 3.1 70B Instruct at $0.23/M input, $0.40/M output.
DeepInfra added Gemma 2 9B IT at $0.0500/M input, $0.0600/M output.
DeepInfra added DeepSeek V3 at $0.20/M input, $0.85/M output.
DeepInfra added Llama 3.1 405B Instruct at $2.70/M input, $8.00/M output.