Price Changelog

Every pricing change, new model, and deprecation — dated and linked to the source. Data is verified against official provider pricing pages. Found an error? Email us and it will be corrected within 48 hours.

October 2026

10-23Price change
o3 · OpenAI · in1m $10.00/M → $20.00/M
Source ↗
10-23Price change
o3 · OpenAI · out1m $40.00/M → $80.00/M
Source ↗
10-23Price change
Claude Opus 5 · Anthropic · in1m $5.00/M → $4.00/M
Source ↗
10-23Price change
Claude Opus 5 · Anthropic · out1m $25.00/M → $20.00/M
Source ↗
10-23Price change
DeepSeek V3 · DeepSeek · in1m $0.27/M → $0.25/M
Source ↗
10-23Price change
DeepSeek V3 · DeepSeek · out1m $1.10/M → $0.95/M
Source ↗
10-23Price change
DeepSeek R1 · DeepSeek · in1m $0.55/M → $0.50/M
Source ↗
10-23Price change
DeepSeek R1 · DeepSeek · out1m $2.19/M → $2.15/M
Source ↗
10-23Price change
DeepSeek V4.1-Flash · DeepSeek · in1m $0.15/M → $0.30/M
Source ↗
10-23Price change
DeepSeek V4.1-Flash · DeepSeek · out1m $0.60/M → $1.20/M
Source ↗
10-23Price change
DeepSeek V4-Pro · DeepSeek · in1m $0.66/M → $1.32/M
Source ↗
10-23Price change
DeepSeek V4-Pro · DeepSeek · out1m $1.98/M → $3.96/M
Source ↗
10-23Price change
Mistral Large 4 · Mistral AI · in1m $0.68/M → $1.36/M
Source ↗
10-23Price change
Mistral Large 4 · Mistral AI · out1m $2.09/M → $4.18/M
Source ↗
10-23Price change
Ministral 8B · Mistral AI · in1m $0.10/M → $0.15/M
Source ↗
10-23Price change
Ministral 8B · Mistral AI · out1m $0.10/M → $0.15/M
Source ↗
10-23Price change
Ministral 3B · Mistral AI · in1m $0.04/M → $0.10/M
Source ↗
10-23Price change
Ministral 3B · Mistral AI · out1m $0.04/M → $0.10/M
Source ↗
10-23New model
GPT-6 Sol · OpenAI
Added GPT-6 Sol. Input: $2/1M, Output: $10/1M. 1M context.
Source ↗
10-23New model
Claude Sonnet 5.5 · Anthropic
Added Claude Sonnet 5.5. Input: $2/1M, Output: $10/1M. 200K context.
Source ↗
10-23New model
Claude Sonnet 5.5 (Batch) · Anthropic
Added Claude Sonnet 5.5 (Batch). Input: $1/1M, Output: $5/1M. 200K context.
Source ↗
10-23New model
Claude Fable 5.1 · Anthropic
Added Claude Fable 5.1. Input: $10/1M, Output: $50/1M. 200K context.
Source ↗
10-23New model
Claude Fable 5.1 (Batch) · Anthropic
Added Claude Fable 5.1 (Batch). Input: $5/1M, Output: $25/1M. 200K context.
Source ↗
10-23New model
Gemini 4 Argon · Google
Added Gemini 4 Argon. Input: $2/1M, Output: $10/1M. 1M context.
Source ↗
10-23New model
Gemini 3.5 Flash · Google
Added Gemini 3.5 Flash. Input: $1.5/1M, Output: $9/1M. 1M context.
Source ↗
10-23New model
Grok Build 0.1 · xAI
Added Grok Build 0.1. Input: $1/1M, Output: $2/1M. 256K context.
Source ↗
10-23Price change
Claude Sonnet 5 · Anthropic · in1m 3 → 2
Permanent price cut: input $3→$2, output $15→$10/1M tokens
Source ↗
10-23Price change
Claude Sonnet 5 (Batch) · Anthropic · in1m 1.5 → 1
Batch pricing updated to match standard price cut: $1/$5 per 1M tokens
Source ↗
10-23Price change
Gemini 3 Flash · Google · inCached1m 0.125 → 0.05
Cached input price reduced from $0.125 to $0.05/1M tokens
Source ↗
10-23status_change
Mistral Large 2 · Mistral AI · status active → deprecated
Superseded by Mistral Large 4
Source ↗
10-23status_change
Mistral Small 3 · Mistral AI · status active → deprecated
Superseded by Mistral Small 4
Source ↗
10-23status_change
Claude Opus 4 · Anthropic · status active → deprecated
Superseded by Claude Opus 5 / 5.5
Source ↗
10-23status_change
Claude Sonnet 4 · Anthropic · status active → deprecated
Superseded by Claude Sonnet 5
Source ↗
10-23New model
GPT-6 Astra · OpenAI
Added GPT-6 Astra — new OpenAI flagship. Input: $10/1M, Output: $50/1M. 1M context.
Source ↗
10-23New model
GPT-6.1 Sol · OpenAI
Added GPT-6.1 Sol. Input: $2/1M, Output: $10/1M. Released Sept 2026.
Source ↗
10-23New model
GPT-6 Luna · OpenAI
Added GPT-6 Luna — cheapest GPT-6. Input: $0.10/1M, Output: $0.50/1M.
Source ↗
10-23New model
GPT-5.6 Terra · OpenAI
Added GPT-5.6 Terra. Input: $2/1M, Output: $12/1M. GA July 2026.
Source ↗
10-23New model
GPT-5.6 Luna · OpenAI
Added GPT-5.6 Luna. Input: $0.20/1M, Output: $1.20/1M.
Source ↗
10-23New model
Claude Opus 5.5 · Anthropic
Added Claude Opus 5.5 — latest Anthropic flagship, Sept 2026. Input: $4/1M, Output: $20/1M.
Source ↗
10-23New model
Claude Haiku 5.5 · Anthropic
Added Claude Haiku 5.5 — cheapest Anthropic model. Input: $0.10/1M (≤100K ctx), Output: $0.50/1M.
Source ↗
10-23New model
DeepSeek V4.1-Flash · DeepSeek
Added DeepSeek V4.1-Flash. Input: $0.15/1M, Output: $0.60/1M (off-peak).
Source ↗
10-23New model
DeepSeek V4-Pro · DeepSeek
Added DeepSeek V4-Pro. Input: $0.66/1M, Output: $1.98/1M (off-peak).
Source ↗
10-23New model
Mistral Large 4 · Mistral AI
Added Mistral Large 4 — new flagship. Input: $0.68/1M, Output: $2.09/1M (preview pricing).
Source ↗
10-23New model
Mistral Medium 3.5 · Mistral AI
Added Mistral Medium 3.5. Input: $1.5/1M, Output: $7.5/1M.
Source ↗
10-23New model
Mistral Small 4 · Mistral AI
Added Mistral Small 4 — successor to Small 3. Input: $0.15/1M, Output: $0.60/1M.
Source ↗
10-23New model
Devstral 2 · Mistral AI
Added Devstral 2 — code specialist. Input: $0.40/1M, Output: $2.00/1M.
Source ↗
10-23New model
Devstral Small 2 · Mistral AI
Added Devstral Small 2. Input: $0.10/1M, Output: $0.30/1M.
Source ↗
10-23New model
Grok 4.7 · xAI
Added Grok 4.7 — xAI latest flagship, 500K context. Input: $2/1M, Output: $6/1M.
Source ↗
10-23New model
Grok 4.3 · xAI
Added Grok 4.3 — 1M context. Input: $1.25/1M, Output: $2.50/1M.
Source ↗
10-23New model
Sonar Reasoning Pro · Perplexity
Added Sonar Reasoning Pro — search-augmented reasoning. Input: $2/1M, Output: $8/1M.
Source ↗
10-23Site update
Bulk price verification: all 130 models verified against official pricing pages. Added 17 new models, updated prices for 5 models, deprecated 4 superseded models.
Source ↗

August 2026

08-19Site update
Launched public pricing changelog, /methodology, /sources, and /data pages
Source ↗
08-19New model
Llama 4 Maverick · Meta
Added Llama 4 Maverick and Llama 4 Scout to dataset. Input: $0.27/1M, Output: $0.85/1M. 1M context.
Source ↗
08-19New model
Llama 4 Scout · Meta
Added to dataset. Input: $0.18/1M, Output: $0.59/1M. 128K context.
Source ↗
08-19New model
DeepSeek V3 · DeepSeek
Added DeepSeek V3 and DeepSeek R1 to dataset. Input: $0.27/1M, Output: $1.1/1M. 64K context.
Source ↗
08-19New model
DeepSeek R1 · DeepSeek
Added to dataset. Input: $0.55/1M, Output: $2.19/1M. 64K context.
Source ↗
08-19New model
DeepSeek R1 Distill Llama 70B · DeepSeek
Added to dataset. Input: $0.88/1M, Output: $0.88/1M. 128K context.
Source ↗
08-19New model
GPT-5.4 · OpenAI
Added to dataset. Input: $2.5/1M, Output: $15.0/1M. 1M context.
Source ↗
08-19New model
GPT-5 mini · OpenAI
Added to dataset. Input: $0.25/1M, Output: $2.0/1M. 400K context.
Source ↗
08-19New model
GPT-4o · OpenAI
Added to dataset. Input: $2.5/1M, Output: $10.0/1M. 128K context.
Source ↗
08-19New model
GPT-4o mini · OpenAI
Added to dataset. Input: $0.15/1M, Output: $0.6/1M. 128K context.
Source ↗
08-19New model
o3 · OpenAI
Added to dataset. Input: $10.0/1M, Output: $40.0/1M. 200K context.
Source ↗
08-19New model
o4-mini · OpenAI
Added to dataset. Input: $1.1/1M, Output: $4.4/1M. 200K context.
Source ↗
08-19New model
text-embedding-3-large · OpenAI
Added to dataset. Input: $0.13/1M, Output: $0.0/1M. 8K context.
Source ↗
08-19New model
GPT-4 Turbo · OpenAI
Added to dataset. Input: $10.0/1M, Output: $30.0/1M. 128K context. (deprecated/previous-gen)
Source ↗
08-19New model
Claude Sonnet 5 · Anthropic
Added to dataset. Input: $3.0/1M, Output: $15.0/1M. 200K context.
Source ↗
08-19New model
Claude Haiku 4.5 · Anthropic
Added to dataset. Input: $1.0/1M, Output: $5.0/1M. 200K context.
Source ↗
08-19New model
Claude Haiku 3.5 · Anthropic
Added to dataset. Input: $0.8/1M, Output: $4.0/1M. 200K context.
Source ↗
08-19New model
Claude 3.5 Sonnet · Anthropic
Added to dataset. Input: $3.0/1M, Output: $15.0/1M. 200K context.
Source ↗
08-19New model
Claude 3 Opus · Anthropic
Added to dataset. Input: $15.0/1M, Output: $75.0/1M. 200K context. (deprecated/previous-gen)
Source ↗
08-19New model
Claude 3 Haiku · Anthropic
Added to dataset. Input: $0.25/1M, Output: $1.25/1M. 200K context. (deprecated/previous-gen)
Source ↗
08-19New model
Gemini 3 Flash · Google
Added Gemini 3 Pro and Gemini 3 Flash. Input: $0.5/1M, Output: $3.0/1M. 1M context.
Source ↗
08-19New model
Gemini 2.5 Pro · Google
Added to dataset. Input: $1.25/1M, Output: $10.0/1M. 2M context.
Source ↗
08-19New model
Gemini 2.5 Flash · Google
Added to dataset. Input: $0.3/1M, Output: $2.5/1M. 1M context.
Source ↗
08-19New model
Gemini 2.5 Flash-Lite · Google
Added to dataset. Input: $0.1/1M, Output: $0.4/1M. 1M context.
Source ↗
08-19New model
Gemini 2.0 Flash · Google
Added to dataset. Input: $0.1/1M, Output: $0.4/1M. 1M context.
Source ↗
08-19New model
Gemini 2.0 Flash-Lite · Google
Added to dataset. Input: $0.075/1M, Output: $0.3/1M. 1M context.
Source ↗
08-19New model
Gemma 3 12B · Google
Added to dataset. Input: $0.12/1M, Output: $0.12/1M. 128K context.
Source ↗
08-19New model
Llama 3.3 70B · Meta
Added to dataset. Input: $0.23/1M, Output: $0.4/1M. 128K context.
Source ↗
08-19New model
Llama 3.1 405B · Meta
Added to dataset. Input: $1.0/1M, Output: $3.0/1M. 128K context.
Source ↗
08-19New model
Llama 3.1 8B · Meta
Added to dataset. Input: $0.05/1M, Output: $0.08/1M. 128K context.
Source ↗
08-19New model
Llama 3.2 3B · Meta
Added to dataset. Input: $0.06/1M, Output: $0.06/1M. 128K context.
Source ↗
08-19New model
Mistral Small 3 · Mistral AI
Added to dataset. Input: $0.1/1M, Output: $0.3/1M. 128K context.
Source ↗
08-19New model
Ministral 8B · Mistral AI
Added to dataset. Input: $0.1/1M, Output: $0.1/1M. 128K context.
Source ↗
08-19New model
Ministral 3B · Mistral AI
Added to dataset. Input: $0.04/1M, Output: $0.04/1M. 128K context.
Source ↗
08-19New model
Mistral Embed · Mistral AI
Added to dataset. Input: $0.1/1M, Output: $0.0/1M. 8K context.
Source ↗
08-19New model
Grok 3 mini · xAI
Added to dataset. Input: $0.3/1M, Output: $0.5/1M. 131K context.
Source ↗
08-19New model
Command R · Cohere
Added to dataset. Input: $0.15/1M, Output: $0.6/1M. 128K context.
Source ↗
08-19New model
Rerank v3 · Cohere
Added to dataset. Input: $2.0/1M, Output: $0.0/1M. 4K context.
Source ↗
08-19New model
Qwen 3 72B · Alibaba
Added to dataset. Input: $0.9/1M, Output: $0.9/1M. 128K context.
Source ↗
08-19New model
Qwen 3 32B · Alibaba
Added to dataset. Input: $0.4/1M, Output: $0.4/1M. 128K context.
Source ↗
08-19New model
QwQ 32B · Alibaba
Added to dataset. Input: $0.9/1M, Output: $0.9/1M. 128K context.
Source ↗
08-19New model
Qwen 3 14B · Alibaba
Added to dataset. Input: $0.15/1M, Output: $0.15/1M. 128K context.
Source ↗