Change feed
What changed, model by model
Every operational change the tracker detects, from the last 30 days: vendor list-price moves, shifts in the cheapest credible provider, context window changes, benchmark movement, and newly tracked models. Detected automatically every 3 hours; how it works is on the methodology page. Subscribe via RSS.
Qwen
- Qwen3.8 27B cheapest credible provider via Alibaba now $0.5 per 1M input tokens (was $0.4, +25%) cheapest provider · Aug 24
- Qwen3.8 27B context window changed from 262,144 to 1,000,000 tokens context · Aug 19
- Qwen3.8 27B cheapest credible provider via Chutes now $3 per 1M output tokens (was $3.2, -6%) cheapest provider · Aug 19
- Qwen3.8 27B cheapest credible provider via Chutes now $0.4 per 1M input tokens (was $0.45, -11%) cheapest provider · Aug 19
- Qwen3.8 27B added to the tracker: $0.45/$3.2 per 1M, 262K context. tracked · Aug 15
Alibaba
- Qwen3.5 397B-A17B cheapest credible provider via Alibaba now $2.34 per 1M output tokens (was $1.925, +22%) cheapest provider · Aug 24
- Qwen3.5 397B-A17B cheapest credible provider via Alibaba now $0.39 per 1M input tokens (was $0.3025, +29%) cheapest provider · Aug 24
- Qwen3.6 27B cheapest credible provider via Chutes now $2 per 1M output tokens (was $2.4, -17%) cheapest provider · Aug 20
- Qwen3.6 27B cheapest credible provider via SiliconFlow now $0.3 per 1M input tokens (was $0.32, -6%) cheapest provider · Aug 17
- Qwen3.8-2.4T-A95B Qwen3.8 2.4T A95B: context window changed from 1,010,000 to 1,048,576 tokens context · Aug 14
- Qwen3.6 27B cheapest credible provider via Phala now $0.32 per 1M input tokens (was $0.289, +11%) cheapest provider · Aug 14
- Qwen3.8-2.4T-A95B Qwen3.8 2.4T A95B: context window changed from 1,000,000 to 1,010,000 tokens context · Aug 14
- Qwen3.8-2.4T-A95B Qwen3.8 2.4T A95B: context window changed from 262,144 to 1,000,000 tokens context · Aug 13
- Qwen3.8-2.4T-A95B Qwen3.8 2.4T A95B (Qwen) added to the tracker: $2/$6 per 1M, 262K context. tracked · Aug 12
- Qwen3.5 397B-A17B cheapest credible provider via DigitalOcean now $0.3025 per 1M input tokens (was $0.385, -21%) cheapest provider · Aug 8
- Qwen3.5 397B-A17B cheapest credible provider via DigitalOcean now $1.925 per 1M output tokens (was $2.45, -21%) cheapest provider · Aug 8
- Qwen3-Coder-Next cheapest credible provider via Parasail now $0.12 per 1M input tokens (was $0.11, +9%) cheapest provider · Aug 7
- Qwen3.5 397B-A17B Humanity's Last Exam improved from 27.3% to 29.0% (+6%) benchmark · Aug 6
- Qwen 3.6 35B-A3B Humanity's Last Exam improved from 20.2% to 22.2% (+10%) benchmark · Aug 6
- Qwen3.5-122B-A10B Humanity's Last Exam improved from 23.4% to 25.2% (+8%) benchmark · Aug 6
- Qwen3.5-27B Humanity's Last Exam improved from 22.2% to 23.9% (+8%) benchmark · Aug 6
- Qwen 3.6 Max (Preview) Humanity's Last Exam improved from 28.9% to 30.8% (+7%) benchmark · Aug 6
- Qwen3-Coder-Next Humanity's Last Exam improved from 9.3% to 10.1% (+9%) benchmark · Aug 6
- Qwen3-Max Humanity's Last Exam improved from 11.1% to 11.9% (+7%) benchmark · Aug 6
- Qwen3-Max-Thinking Humanity's Last Exam improved from 26.2% to 28.0% (+7%) benchmark · Aug 6
- Qwen 3.7 Max Humanity's Last Exam improved from 38.1% to 40.5% (+6%) benchmark · Aug 6
- Qwen 3.6 Plus Humanity's Last Exam improved from 25.7% to 27.8% (+8%) benchmark · Aug 6
- Qwen3.5-Omni-Flash Humanity's Last Exam improved from 7.1% to 7.6% (+7%) benchmark · Aug 6
- Qwen3.5-Omni-Plus Humanity's Last Exam improved from 13.9% to 14.9% (+7%) benchmark · Aug 6
- Qwen3.5-9B Humanity's Last Exam improved from 13.3% to 14.9% (+12%) benchmark · Aug 6
- Qwen3.7-Plus Humanity's Last Exam improved from 33.4% to 35.6% (+7%) benchmark · Aug 6
- Qwen3.7-Max Humanity's Last Exam improved from 38.1% to 40.5% (+6%) benchmark · Aug 6
- Qwen 3.8 Max SciCode improved from 45.0% to 52.9% (+18%) benchmark · Aug 6
- Qwen 3.6 35B-A3B cheapest credible provider via DeepInfra now $0.95 per 1M output tokens (was $1, -5%) cheapest provider · Jul 31
- Qwen 3.7 Max cheapest credible provider via Alibaba now $1.475 per 1M input tokens (was $1.25, +18%) cheapest provider · Jul 31
- Qwen 3.7 Max cheapest credible provider via Alibaba now $4.425 per 1M output tokens (was $3.75, +18%) cheapest provider · Jul 31
- Qwen3-Coder-Next cheapest credible provider via Ionstream now $0.11 per 1M input tokens (was $0.12, -8%) cheapest provider · Jul 31
- Qwen 3.6 Max (Preview) vendor list price cut from $6.24 to $6.162 per 1M output tokens (-1%) list price · Jul 30
- Qwen 3.6 Max (Preview) vendor list price cut from $1.04 to $1.027 per 1M input tokens (-1%) list price · Jul 30
- Qwen3.5-27B cheapest credible provider via Alibaba now $1.56 per 1M output tokens (was $2, -22%) cheapest provider · Jul 30
- Qwen3.5-27B cheapest credible provider via Alibaba now $0.195 per 1M input tokens (was $0.26, -25%) cheapest provider · Jul 30
- Qwen 3.6 35B-A3B cheapest credible provider via Venice now $0.1 per 1M input tokens (was $0.14, -29%) cheapest provider · Jul 29
- Qwen3.7 Flash added to the tracker: $0.03/$0.13 per 1M, 1M context. tracked · Jul 28
- Qwen 3.6 35B-A3B cheapest credible provider via AkashML now $0.14 per 1M input tokens (was $0.13, +8%) cheapest provider · Jul 26
- Qwen3.5 397B-A17B cheapest credible provider via DigitalOcean now $2.45 per 1M output tokens (was $3, -18%) cheapest provider · Jul 26
OpenAI
- GPT-5.6 Sol (batch) vendor list price cut from $15 to $10 per 1M output tokens (-33%) list price · Aug 23
- GPT-5.6 Sol (batch) vendor list price cut from $2.5 to $2 per 1M input tokens (-20%) list price · Aug 23
- GPT-5.6 Sol Pro (batch) vendor list price cut from $2.5 to $2 per 1M input tokens (-20%) list price · Aug 23
- GPT-5.6 Sol Pro (batch) vendor list price cut from $15 to $10 per 1M output tokens (-33%) list price · Aug 23
- GPT-5.6 Sol vendor list price cut from $30 to $20 per 1M output tokens (-33%) list price · Aug 23
- GPT-5.6 Sol vendor list price cut from $5 to $4 per 1M input tokens (-20%) list price · Aug 23
- GPT-5.6 Sol Pro vendor list price cut from $30 to $20 per 1M output tokens (-33%) list price · Aug 23
- GPT-5.6 Sol Pro vendor list price cut from $5 to $4 per 1M input tokens (-20%) list price · Aug 23
- GPT-5.6 Terra cheapest credible provider via Azure now $12 per 1M output tokens (was $13.2, -9%) cheapest provider · Aug 13
- GPT-5.6 Terra cheapest credible provider via Azure now $2 per 1M input tokens (was $2.2, -9%) cheapest provider · Aug 13
- GPT-5.6 Terra Pro cheapest credible provider via Azure now $12 per 1M output tokens (was $15, -20%) cheapest provider · Aug 12
- GPT-5.6 Terra Pro cheapest credible provider via Azure now $2 per 1M input tokens (was $2.5, -20%) cheapest provider · Aug 12
- GPT-5.6 Luna cheapest credible provider via Azure now $1.2 per 1M output tokens (was $1.32, -9%) cheapest provider · Aug 12
- GPT-5.6 Luna cheapest credible provider via Azure now $0.2 per 1M input tokens (was $0.22, -9%) cheapest provider · Aug 12
- GPT-5.6 Luna Pro cheapest credible provider via Azure now $1.2 per 1M output tokens (was $6, -80%) cheapest provider · Aug 12
- GPT-5.6 Luna Pro cheapest credible provider via Azure now $0.2 per 1M input tokens (was $1, -80%) cheapest provider · Aug 12
- GPT-5.4 Mini Humanity's Last Exam improved from 26.6% to 28.1% (+6%) benchmark · Aug 6
- GPT-5.4 Nano Humanity's Last Exam improved from 26.5% to 28.3% (+7%) benchmark · Aug 6
- gpt-oss-120b Humanity's Last Exam improved from 18.5% to 19.6% (+6%) benchmark · Aug 6
- gpt-oss-20b Humanity's Last Exam improved from 9.8% to 11.0% (+12%) benchmark · Aug 6
- GPT-5 nano Humanity's Last Exam improved from 8.2% to 9.5% (+16%) benchmark · Aug 6
- GPT-5.6 Luna Humanity's Last Exam improved from 37.2% to 39.5% (+6%) benchmark · Aug 6
- GPT-5.1-Codex Humanity's Last Exam improved from 23.4% to 25.7% (+10%) benchmark · Aug 6
- GPT-5.3 Codex Humanity's Last Exam improved from 39.9% to 42.5% (+7%) benchmark · Aug 6
- GPT‑5-Codex Humanity's Last Exam improved from 25.6% to 27.8% (+9%) benchmark · Aug 6
- GPT-5.2 Codex Humanity's Last Exam improved from 33.5% to 35.7% (+7%) benchmark · Aug 6
- GPT-5.1-Codex-mini Humanity's Last Exam improved from 16.9% to 18.5% (+9%) benchmark · Aug 6
- GPT-5.5 Instant Humanity's Last Exam improved from 20.3% to 21.6% (+6%) benchmark · Aug 6
- GPT-5.6 Sol (batch) added to the tracker: $2.5/$15 per 1M, 1.1M context. tracked · Aug 6
- GPT-5.6 Sol Pro (batch) added to the tracker: $2.5/$15 per 1M, 1.1M context. tracked · Aug 6
- GPT-5.6 Terra (batch) added to the tracker: $1/$6 per 1M, 1.1M context. tracked · Aug 6
- GPT-5.6 Terra Pro (batch) added to the tracker: $1/$6 per 1M, 1.1M context. tracked · Aug 6
- GPT-5.6 Luna (batch) added to the tracker: $0.1/$0.6 per 1M, 1.1M context. tracked · Aug 6
- GPT-5.6 Luna Pro (batch) added to the tracker: $0.1/$0.6 per 1M, 1.1M context. tracked · Aug 6
- GPT-5.2 ARC-AGI-2 dropped from 72.9% to 52.9% (-27%) benchmark · Aug 5
- GPT-5.2 ARC-AGI dropped from 94.5% to 86.2% (-9%) benchmark · Aug 5
- GPT-5.2 ARC-AGI-2 improved from 52.9% to 72.9% (+38%) benchmark · Aug 5
- GPT-5.2 ARC-AGI improved from 86.2% to 94.5% (+10%) benchmark · Aug 5
- GPT-5.6 Terra Pro vendor list price cut from $15 to $12 per 1M output tokens (-20%) list price · Aug 1
- GPT-5.6 Terra Pro vendor list price cut from $2.5 to $2 per 1M input tokens (-20%) list price · Aug 1
- GPT-5.6 Luna vendor list price cut from $6 to $1.2 per 1M output tokens (-80%) list price · Aug 1
- GPT-5.6 Luna cheapest credible provider via Amazon Bedrock now $1.32 per 1M output tokens (was $6, -78%) cheapest provider · Aug 1
- GPT-5.6 Luna cheapest credible provider via Amazon Bedrock now $0.22 per 1M input tokens (was $1, -78%) cheapest provider · Aug 1
- GPT-5.6 Luna vendor list price cut from $1 to $0.2 per 1M input tokens (-80%) list price · Aug 1
- GPT-5.6 Luna Pro vendor list price cut from $6 to $1.2 per 1M output tokens (-80%) list price · Aug 1
- GPT-5.6 Luna Pro vendor list price cut from $1 to $0.2 per 1M input tokens (-80%) list price · Aug 1
- GPT-5.6 Terra vendor list price cut from $2.5 to $2 per 1M input tokens (-20%) list price · Aug 1
- GPT-5.6 Terra cheapest credible provider via Amazon Bedrock now $13.2 per 1M output tokens (was $15, -12%) cheapest provider · Aug 1
- GPT-5.6 Terra cheapest credible provider via Amazon Bedrock now $2.2 per 1M input tokens (was $2.5, -12%) cheapest provider · Aug 1
- GPT-5.6 Terra vendor list price cut from $15 to $12 per 1M output tokens (-20%) list price · Aug 1
NVIDIA
- Nemotron 3.5 Lightning cheapest credible provider via DeepInfra now $0.2 per 1M output tokens (was $0.25, -20%) cheapest provider · Aug 23
- Nemotron 3.5 Lightning cheapest credible provider via DeepInfra now $0.08 per 1M input tokens (was $0.1, -20%) cheapest provider · Aug 23
- Nemotron 3.5 Lightning context window changed from 1,000,000 to 262,144 tokens context · Aug 20
- Nemotron 3.5 Lightning context window changed from 1,048,576 to 1,000,000 tokens context · Aug 13
- Nemotron 3.5 Lightning context window changed from 1,000,000 to 1,048,576 tokens context · Aug 13
- Nemotron 3.5 Lightning context window changed from 262,144 to 1,000,000 tokens context · Aug 13
- Nemotron 3.5 Lightning added to the tracker: $0.1/$0.25 per 1M, 262K context. tracked · Aug 11
- Nemotron 3-Nano-30B-A3B cheapest credible provider via Crusoe now $0.2 per 1M output tokens (was $0.24, -17%) cheapest provider · Jul 26
- Nemotron 3-Nano-30B-A3B cheapest credible provider via Crusoe now $0.05 per 1M input tokens (was $0.06, -17%) cheapest provider · Jul 26
Moonshot
- Kimi K2.5 cheapest credible provider via DigitalOcean now $0.5 per 1M input tokens (was $0.375, +33%) cheapest provider · Aug 23
- Kimi K2.5 cheapest credible provider via DigitalOcean now $2.7 per 1M output tokens (was $2.025, +33%) cheapest provider · Aug 23
- Kimi K2.7 Code vendor list price cut from $8 to $4 per 1M output tokens (-50%) list price · Aug 15
- Kimi K2.7 Code vendor list price cut from $1.9 to $0.95 per 1M input tokens (-50%) list price · Aug 15
- Kimi K2.7 Code cheapest credible provider via Ambient now $0.69 per 1M input tokens (was $0.8592, -20%) cheapest provider · Aug 14
- Kimi K2.7 Code cheapest credible provider via Ambient now $3.49 per 1M output tokens (was $3.8, -8%) cheapest provider · Aug 14
- Kimi K3 cheapest credible provider via Morph now $14 per 1M output tokens (was $15, -7%) cheapest provider · Aug 10
- Kimi K3 cheapest credible provider via Morph now $2.8 per 1M input tokens (was $3, -7%) cheapest provider · Aug 10
- Kimi K2.6 cheapest credible provider via Crusoe now $3.5 per 1M output tokens (was $3.2, +9%) cheapest provider · Aug 9
- Kimi K2.6 cheapest credible provider via Crusoe now $0.7 per 1M input tokens (was $0.76, -8%) cheapest provider · Aug 9
- Kimi K3 cheapest credible provider via DigitalOcean now $15 per 1M output tokens (was $14, +7%) cheapest provider · Aug 7
- Kimi K2.7 Code Humanity's Last Exam improved from 32.8% to 35.0% (+7%) benchmark · Aug 6
- Kimi K3 Humanity's Last Exam improved from 44.3% to 46.9% (+6%) benchmark · Aug 6
- Kimi K2 Thinking Humanity's Last Exam improved from 22.3% to 23.8% (+7%) benchmark · Aug 6
- Kimi K2 Humanity's Last Exam improved from 7.0% to 7.4% (+6%) benchmark · Aug 6
- Kimi K2.6 cheapest credible provider via DigitalOcean now $3.2 per 1M output tokens (was $3.5, -9%) cheapest provider · Aug 6
- Kimi K2.6 cheapest credible provider via DigitalOcean now $0.76 per 1M input tokens (was $0.7, +9%) cheapest provider · Aug 6
- Kimi K3 cheapest credible provider via Morph now $14 per 1M output tokens (was $15, -7%) cheapest provider · Aug 4
- Kimi K2.5 cheapest credible provider via DigitalOcean now $2.025 per 1M output tokens (was $3, -32%) cheapest provider · Jul 26
- Kimi K2.5 cheapest credible provider via DigitalOcean now $0.375 per 1M input tokens (was $0.6, -38%) cheapest provider · Jul 26
Tencent
- Hy-MT2-7B added to the tracker: $0.074/$0.295 per 1M, 8K context. tracked · Aug 23
- Hy-MT2-30B-A3B added to the tracker: $0.074/$0.295 per 1M, 8K context. tracked · Aug 20
- Hy-MT2-1.8B added to the tracker: $0.044/$0.177 per 1M, 8K context. tracked · Aug 20
- Hy3 Humanity's Last Exam improved from 31.6% to 33.5% (+6%) benchmark · Aug 6
Z.ai (Zhipu AI)
- GLM-4.7 cheapest credible provider via Google now $0.6 per 1M input tokens (was $0.52, +15%) cheapest provider · Aug 22
- GLM-4.7 cheapest credible provider via Google now $2.2 per 1M output tokens (was $1.85, +19%) cheapest provider · Aug 22
- GLM-5.1 cheapest credible provider via Chutes now $3.08 per 1M output tokens (was $4.3, -28%) cheapest provider · Aug 20
- GLM-5.1 cheapest credible provider via DigitalOcean now $4.3 per 1M output tokens (was $3.74, +15%) cheapest provider · Aug 15
- GLM-5.1 cheapest credible provider via DigitalOcean now $0.975 per 1M input tokens (was $1.19, -18%) cheapest provider · Aug 15
- GLM-4.7 cheapest credible provider via AtlasCloud now $1.85 per 1M output tokens (was $2.2, -16%) cheapest provider · Aug 15
- GLM-4.7 cheapest credible provider via AtlasCloud now $0.52 per 1M input tokens (was $0.6, -13%) cheapest provider · Aug 15
- GLM-5.2 cheapest credible provider via Sail Research now $0.5 per 1M input tokens (was $0.9, -44%) cheapest provider · Aug 13
- GLM-4.7 cheapest credible provider via Google now $2.2 per 1M output tokens (was $1.85, +19%) cheapest provider · Aug 12
- GLM-4.7 cheapest credible provider via Google now $0.6 per 1M input tokens (was $0.52, +15%) cheapest provider · Aug 12
- GLM-5.1 cheapest credible provider via SiliconFlow now $3.74 per 1M output tokens (was $4.3, -13%) cheapest provider · Aug 11
- GLM-5.1 cheapest credible provider via SiliconFlow now $1.19 per 1M input tokens (was $0.98, +21%) cheapest provider · Aug 11
- GLM-5.1 reported GPQA Diamond score rose from 85.5% to 89.9% (+5%) benchmark · Aug 10
- GLM-5.2 cheapest credible provider via Sail Research now $0.9 per 1M input tokens (was $0.966, -7%) cheapest provider · Aug 8
- GLM-4.7 Humanity's Last Exam improved from 25.1% to 27.4% (+9%) benchmark · Aug 6
- GLM-5 Humanity's Last Exam improved from 27.2% to 29.3% (+8%) benchmark · Aug 6
- GLM-5.1 Humanity's Last Exam improved from 28.0% to 30.1% (+7%) benchmark · Aug 6
- GLM-4.7 cheapest credible provider via AtlasCloud now $1.85 per 1M output tokens (was $2.2, -16%) cheapest provider · Aug 6
- GLM-4.7 cheapest credible provider via AtlasCloud now $0.52 per 1M input tokens (was $0.6, -13%) cheapest provider · Aug 6
- GLM-5.2 cheapest credible provider via Alibaba now $3.036 per 1M output tokens (was $4.1, -26%) cheapest provider · Jul 31
- GLM-5.2 cheapest credible provider via Alibaba now $0.966 per 1M input tokens (was $1.1, -12%) cheapest provider · Jul 31
- GLM-5.1 cheapest credible provider via DigitalOcean now $4.3 per 1M output tokens (was $3.08, +40%) cheapest provider · Jul 30
- GLM-4.7 APEX improved from 3.1% to 8.7% (+181%) benchmark · Jul 30
- GLM-4.7 cheapest credible provider via Google now $0.6 per 1M input tokens (was $0.52, +15%) cheapest provider · Jul 29
- GLM-4.7 cheapest credible provider via Google now $2.2 per 1M output tokens (was $1.85, +19%) cheapest provider · Jul 29
- GLM-4.7 cheapest credible provider via AtlasCloud now $0.52 per 1M input tokens (was $0.6, -13%) cheapest provider · Jul 27
- GLM-4.7 cheapest credible provider via AtlasCloud now $1.85 per 1M output tokens (was $2.2, -16%) cheapest provider · Jul 27
Meta
- Muse Spark 1.2 Contributor added to the tracker: $0.1/$0.2 per 1M, 1M context. tracked · Aug 22
- Muse Glimmer 30B cheapest credible provider via Phala now $1.1 per 1M output tokens (was $1.2, -8%) cheapest provider · Aug 21
- Muse Glimmer 30B cheapest credible provider via DeepInfra now $0.3 per 1M input tokens (was $0.35, -14%) cheapest provider · Aug 13
- Muse Glimmer 30B cheapest credible provider via DeepInfra now $1.2 per 1M output tokens (was $1.5, -20%) cheapest provider · Aug 13
- Muse Glimmer 30B added to the tracker: $0.35/$1.5 per 1M, 131K context. tracked · Aug 11
MiniMax
- MiniMax-M2.5 cheapest credible provider via Venice now $0.27 per 1M input tokens (was $0.22, +23%) cheapest provider · Aug 21
- MiniMax-M2.5 cheapest credible provider via Venice now $0.95 per 1M output tokens (was $0.9, +6%) cheapest provider · Aug 21
- MiniMax M3 (batch) cheapest credible provider via Together now $1.2 per 1M output tokens (was $0.6, +100%) cheapest provider · Aug 20
- MiniMax M3 (batch) cheapest credible provider via Together now $0.3 per 1M input tokens (was $0.15, +100%) cheapest provider · Aug 20
- MiniMax-M2.5 vendor list price cut from $2.4 to $1.2 per 1M output tokens (-50%) list price · Aug 15
- MiniMax-M2.5 vendor list price cut from $0.6 to $0.3 per 1M input tokens (-50%) list price · Aug 15
- MiniMax-M2.7 vendor list price cut from $2.4 to $1.2 per 1M output tokens (-50%) list price · Aug 14
- MiniMax-M2.7 vendor list price cut from $0.6 to $0.3 per 1M input tokens (-50%) list price · Aug 14
- MiniMax-M3 cheapest credible provider via DeepInfra now $1.1 per 1M output tokens (was $1.2, -8%) cheapest provider · Aug 12
- MiniMax-M3 cheapest credible provider via DeepInfra now $0.28 per 1M input tokens (was $0.3, -7%) cheapest provider · Aug 12
- MiniMax-M2.5 cheapest credible provider via Inceptron now $0.22 per 1M input tokens (was $0.15, +47%) cheapest provider · Aug 8
- MiniMax-M3 Humanity's Last Exam improved from 37.1% to 39.0% (+5%) benchmark · Aug 6
- MiniMax-M2.5 Humanity's Last Exam improved from 19.1% to 20.5% (+7%) benchmark · Aug 6
- MiniMax-M2 Humanity's Last Exam improved from 12.5% to 13.7% (+10%) benchmark · Aug 6
- MiniMax-M2.7 Humanity's Last Exam improved from 28.1% to 29.6% (+5%) benchmark · Aug 6
- MiniMax-M2 vendor list price raised from $0.255 to $0.3 per 1M input tokens (+18%) list price · Jul 31
- MiniMax-M2 vendor list price raised from $1.02 to $1.2 per 1M output tokens (+18%) list price · Jul 31
- MiniMax M3 (batch) added to the tracker: $0.15/$0.6 per 1M, 524K context. tracked · Jul 28
DeepSeek
- DeepSeek V4 Flash Vision Exp added to the tracker: $0.22/$0.66 per 1M, 1M context. tracked · Aug 21
- DeepSeek-V4-Pro reported WebDev Arena score fell from 1584.45 to 1463.88 (-8%) benchmark · Aug 19
- DeepSeek-V4-Pro reported WeirdML score fell from 66.2% to 48.9% (-26%) benchmark · Aug 19
- DeepSeek-V4-Pro reported WebDev Arena score rose from 1463.88 to 1584.45 (+8%) benchmark · Aug 18
- DeepSeek-V4-Pro reported WeirdML score rose from 48.9% to 66.2% (+35%) benchmark · Aug 18
- DeepSeek-V4-Pro-0813 vendor list price raised from $0.87 to $1.98 per 1M output tokens (+128%) list price · Aug 18
- DeepSeek-V4-Pro-0813 cheapest credible provider via DeepSeek now $1.98 per 1M output tokens (was $0.87, +128%) cheapest provider · Aug 18
- DeepSeek-V4-Pro-0813 cheapest credible provider via DeepSeek now $0.66 per 1M input tokens (was $0.435, +52%) cheapest provider · Aug 18
- DeepSeek-V4-Pro-0813 vendor list price raised from $0.435 to $0.66 per 1M input tokens (+52%) list price · Aug 18
- DeepSeek V4 Flash 0731 cheapest credible provider via DeepInfra now $0.18 per 1M output tokens (was $0.26, -31%) cheapest provider · Aug 17
- DeepSeek V4 Flash 0731 context window changed from 1,048,576 to 1,310,720 tokens context · Aug 17
- DeepSeek-V4-Flash cheapest credible provider via DigitalOcean now $0.0679 per 1M input tokens (was $0.084, -19%) cheapest provider · Aug 12
- DeepSeek-V4-Pro-0813 DeepSeek V4 Pro 0813 (DeepSeek) added to the tracker: $0.435/$0.87 per 1M, 1M context. tracked · Aug 12
- DeepSeek V4 Flash 0731 cheapest credible provider via DigitalOcean now $0.08 per 1M input tokens (was $0.13, -38%) cheapest provider · Aug 10
- DeepSeek V4 Flash 0731 cheapest credible provider via BaseTen now $0.13 per 1M input tokens (was $0.14, -7%) cheapest provider · Aug 7
- DeepSeek V4 Flash 0731 cheapest credible provider via BaseTen now $0.26 per 1M output tokens (was $0.28, -7%) cheapest provider · Aug 7
- DeepSeek-V4-Flash cheapest credible provider via DigitalOcean now $0.168 per 1M output tokens (was $0.28, -40%) cheapest provider · Aug 7
- DeepSeek-V4-Flash cheapest credible provider via DigitalOcean now $0.084 per 1M input tokens (was $0.13, -35%) cheapest provider · Aug 7
- DeepSeek-V3.2 Humanity's Last Exam improved from 22.2% to 24.6% (+11%) benchmark · Aug 6
- DeepSeek-V3.2-Exp Humanity's Last Exam improved from 13.8% to 14.9% (+8%) benchmark · Aug 6
- DeepSeek-V4-Flash Humanity's Last Exam improved from 32.1% to 34.8% (+8%) benchmark · Aug 6
- DeepSeek-V3.1 Humanity's Last Exam improved from 6.3% to 6.7% (+6%) benchmark · Aug 6
- DeepSeek-V3.1-Terminus Humanity's Last Exam improved from 15.2% to 16.4% (+8%) benchmark · Aug 6
- DeepSeek-V3.2 cheapest credible provider via DigitalOcean now $0.8 per 1M output tokens (was $0.42, +90%) cheapest provider · Aug 6
- DeepSeek V4 Flash 0731 added to the tracker: $0.14/$0.28 per 1M, 1M context. tracked · Jul 31
- DeepSeek-V3.2 cheapest credible provider via SiliconFlow now $0.42 per 1M output tokens (was $0.378, +11%) cheapest provider · Jul 29
- DeepSeek-V4-Flash cheapest credible provider via SiliconFlow now $0.28 per 1M output tokens (was $0.1932, +45%) cheapest provider · Jul 26
- DeepSeek-V4-Flash cheapest credible provider via SiliconFlow now $0.13 per 1M input tokens (was $0.0966, +35%) cheapest provider · Jul 26
- Gemini 3.7 Flash (batch) vendor list price raised from $1.875 to $3.75 per 1M output tokens (+100%) list price · Aug 21
- Gemini 3.7 Flash (batch) vendor list price raised from $0.375 to $0.75 per 1M input tokens (+100%) list price · Aug 21
- Gemini 3.6 Flash (batch) vendor list price cut from $3.75 to $1.875 per 1M output tokens (-50%) list price · Aug 15
- Gemini 3.6 Flash (batch) cheapest credible provider via Google now $1.875 per 1M output tokens (was $3.75, -50%) cheapest provider · Aug 15
- Gemini 3.6 Flash (batch) cheapest credible provider via Google now $0.375 per 1M input tokens (was $0.75, -50%) cheapest provider · Aug 15
- Gemini 3.6 Flash (batch) vendor list price cut from $0.75 to $0.375 per 1M input tokens (-50%) list price · Aug 15
- Gemini 3.7 Flash (batch) added to the tracker: $0.1875/$0.9375 per 1M, 1M context. tracked · Aug 13
- Gemini 3.5 Flash Lite (batch) added to the tracker: $0.15/$1.25 per 1M, 1M context. tracked · Jul 28
- Gemini 3.6 Flash (batch) added to the tracker: $0.75/$3.75 per 1M, 1M context. tracked · Jul 28
Google DeepMind
- Gemini 3.7 Flash vendor list price raised from $3.75 to $7.5 per 1M output tokens (+100%) list price · Aug 21
- Gemini 3.7 Flash vendor list price raised from $0.75 to $1.5 per 1M input tokens (+100%) list price · Aug 21
- Gemma 4 31B IT cheapest credible provider via OpenInference now $0.08 per 1M input tokens (was $0.1, -20%) cheapest provider · Aug 17
- Gemini 3.6 Flash vendor list price cut from $7.5 to $3.75 per 1M output tokens (-50%) list price · Aug 16
- Gemini 3.6 Flash cheapest credible provider via Google now $3.75 per 1M output tokens (was $7.5, -50%) cheapest provider · Aug 16
- Gemini 3.6 Flash cheapest credible provider via Google now $0.75 per 1M input tokens (was $1.5, -50%) cheapest provider · Aug 16
- Gemini 3.6 Flash vendor list price cut from $1.5 to $0.75 per 1M input tokens (-50%) list price · Aug 16
- Gemma 4 31B IT cheapest credible provider via CoreWeave now $0.1 per 1M input tokens (was $0.08, +25%) cheapest provider · Aug 14
- Gemini 3.7 Flash added to the tracker: $0.375/$1.875 per 1M, 1M context. tracked · Aug 13
- Gemma 4 31B IT cheapest credible provider via OpenInference now $0.08 per 1M input tokens (was $0.1, -20%) cheapest provider · Aug 10
- Gemini 3 Flash GPQA Diamond improved from 83.2% to 89.4% (+7%) benchmark · Aug 7
- Gemini 2.5 Flash-Lite Humanity's Last Exam improved from 6.4% to 6.8% (+6%) benchmark · Aug 6
- Gemma 4 26B A4B Humanity's Last Exam improved from 18.3% to 19.3% (+5%) benchmark · Aug 6
- Gemini 3.6 Flash Humanity's Last Exam improved from 38.3% to 40.8% (+7%) benchmark · Aug 6
- Gemini 3.5 Flash-Lite Humanity's Last Exam improved from 17.5% to 18.8% (+7%) benchmark · Aug 6
- Gemini 3 Pro ARC-AGI-2 dropped from 54.0% to 31.1% (-42%) benchmark · Aug 5
- Gemini 3 Pro ARC-AGI-2 improved from 31.1% to 54.0% (+74%) benchmark · Aug 5
- Gemini 3 Flash ARC-AGI improved from 21.5% to 84.7% (+294%) benchmark · Aug 5
- Gemini 3 Pro APEX improved from 18.4% to 31.5% (+71%) benchmark · Jul 30
Z.ai
- GLM 5.2 (batch) context window changed from 512,000 to 1,048,575 tokens context · Aug 21
- GLM 5.2 (batch) cheapest credible provider via Together now $4.4 per 1M output tokens (was $2.2, +100%) cheapest provider · Aug 20
- GLM 5.2 (batch) cheapest credible provider via Together now $1.4 per 1M input tokens (was $0.7, +100%) cheapest provider · Aug 20
- GLM 5.2 (batch) added to the tracker: $0.7/$2.2 per 1M, 512K context. tracked · Aug 6
Thinking Machines
- Inkling-Small context window changed from 524,288 to 1,048,576 tokens context · Aug 20
- Inkling (batch) cheapest credible provider via Together now $4.05 per 1M output tokens (was $2.025, +100%) cheapest provider · Aug 20
- Inkling (batch) cheapest credible provider via Together now $1 per 1M input tokens (was $0.5, +100%) cheapest provider · Aug 20
- Inkling Humanity's Last Exam improved from 29.7% to 31.9% (+7%) benchmark · Aug 6
- Inkling-Small Humanity's Last Exam improved from 31.6% to 33.3% (+5%) benchmark · Aug 6
- Inkling cheapest credible provider via DeepInfra now $0.95 per 1M input tokens (was $1, -5%) cheapest provider · Aug 6
- Inkling (batch) added to the tracker: $0.5/$2.025 per 1M, 524K context. tracked · Aug 6
- Inkling-Small cheapest credible provider via DeepInfra now $0.45 per 1M input tokens (was $0.58, -22%) cheapest provider · Aug 4
- Inkling-Small Inkling Small: cheapest credible provider via DeepInfra now $1.2 per 1M output tokens (was $1.44, -17%) cheapest provider · Aug 3
- Inkling-Small Inkling Small (Thinking Machines) added to the tracker: $0.58/$1.44 per 1M, 524K context. tracked · Jul 31
Other
- Ox Alpha added to the tracker: 1M context. tracked · Aug 20
- Ling-3.0-flash cheapest credible provider via DeepInfra now $0.18 per 1M output tokens (was $0.22, -18%) cheapest provider · Aug 11
- Ling-3.0-flash cheapest credible provider via DeepInfra now $0.06 per 1M input tokens (was $0.075, -20%) cheapest provider · Aug 11
- Ling-3.0-flash Humanity's Last Exam improved from 22.1% to 23.7% (+7%) benchmark · Aug 6
- Ling-3.0-flash context window changed from 131,072 to 262,144 tokens context · Aug 6
- Ling-3.0-flash context window changed from 262,144 to 131,072 tokens context · Aug 6
- Claude Opus 5 (batch) added to the tracker: $2.5/$12.5 per 1M, 1M context. tracked · Jul 28
MoonshotAI
- Kimi K2.7 Code (batch) cheapest credible provider via Together now $4 per 1M output tokens (was $2, +100%) cheapest provider · Aug 20
- Kimi K2.7 Code (batch) cheapest credible provider via Together now $0.95 per 1M input tokens (was $0.475, +100%) cheapest provider · Aug 20
- Kimi K2.7 Code (batch) added to the tracker: $0.475/$2 per 1M, 262K context. tracked · Aug 6
Motif Technologies
xAI
- Grok 4.6 added to the tracker: $2/$6 per 1M, 500K context. tracked · Aug 12
- Grok 4.5 Humanity's Last Exam improved from 40.3% to 42.7% (+6%) benchmark · Aug 6
- Grok 4.1 Fast Humanity's Last Exam improved from 17.6% to 19.3% (+10%) benchmark · Aug 6
- Grok 4 Fast Humanity's Last Exam improved from 17.0% to 19.1% (+12%) benchmark · Aug 6
- Grok 4 Humanity's Last Exam improved from 23.9% to 26.7% (+12%) benchmark · Aug 6
- Grok 4 ARC-AGI-2 dropped from 29.4% to 16.0% (-46%) benchmark · Aug 5
- Grok 4 ARC-AGI dropped from 79.6% to 66.7% (-16%) benchmark · Aug 5
- Grok 4 ARC-AGI-2 improved from 16.0% to 29.4% (+84%) benchmark · Aug 5
- Grok 4 ARC-AGI improved from 66.7% to 79.6% (+19%) benchmark · Aug 5
ByteDance Seed
- Seed 2.1 Turbo added to the tracker: $0.5/$2.5 per 1M, 262K context. tracked · Aug 12
- Seed-2.0-Code added to the tracker: $0.5/$3 per 1M, 262K context. tracked · Aug 11
Sakana
- Sakana Namazu added to the tracker: $0.95/$4 per 1M, 262K context. tracked · Aug 11
- Namazu added to the tracker: $0.95/$4 per 1M, 262K context. tracked · Aug 11
Upstage
- Solar Pro 4 added to the tracker: $0.03/$0.12 per 1M, 524K context. tracked · Aug 11
- Solar Open 100B Humanity's Last Exam improved from 9.2% to 10.4% (+13%) benchmark · Aug 6
- Solar Pro 3 context window changed from 128,000 to 131,072 tokens context · Aug 5
Anthropic
- Claude Fable 5 GPQA Diamond improved from 78.8% to 85.9% (+9%) benchmark · Aug 7
- Claude Opus 4.8 Humanity's Last Exam improved from 45.7% to 48.7% (+7%) benchmark · Aug 6
- Claude Sonnet 4.6 Humanity's Last Exam improved from 30.0% to 33.6% (+12%) benchmark · Aug 6
- Claude Opus 4.7 Terminal-Bench dropped from 90.2% to 80.2% (-11%) benchmark · Aug 5
- Claude Opus 4.5 APEX improved from 18.4% to 20.7% (+12%) benchmark · Jul 30
- Claude Fable 5 (batch) added to the tracker: $5/$25 per 1M, 1M context. tracked · Jul 28
- Claude Sonnet 5 (batch) added to the tracker: $1/$5 per 1M, 1M context. tracked · Jul 28
Mistral AI
- Mistral Medium 3.5 Humanity's Last Exam improved from 12.8% to 13.8% (+8%) benchmark · Aug 6
- Magistral Medium 1.2 Humanity's Last Exam improved from 9.6% to 10.3% (+7%) benchmark · Aug 6
Z.ai (Zhipu AI),Tsinghua University
Ant Group
- Ring-2.6-1T Humanity's Last Exam improved from 18.3% to 21.6% (+18%) benchmark · Aug 6
Cohere,Cohere Labs (formerly Cohere for AI)
- Command A+ Humanity's Last Exam improved from 11.4% to 12.0% (+5%) benchmark · Aug 6
StepFun
- Step 3.7 Flash Humanity's Last Exam improved from 19.9% to 21.4% (+8%) benchmark · Aug 6
LG AI Research
- EXAONE 4.0 (32B) Humanity's Last Exam improved from 10.5% to 11.4% (+9%) benchmark · Aug 6
- K-EXAONE Humanity's Last Exam improved from 13.1% to 13.9% (+6%) benchmark · Aug 6
Xiaomi Corp
- MiMo-V2-Pro Humanity's Last Exam improved from 28.3% to 30.4% (+7%) benchmark · Aug 6
- MiMo-V2.5-Pro Humanity's Last Exam improved from 33.8% to 35.7% (+6%) benchmark · Aug 6
Arcee AI
- Trinity Large Thinking Humanity's Last Exam improved from 14.7% to 15.8% (+7%) benchmark · Aug 6
Deep Cogito
- Cogito v2.1 Humanity's Last Exam improved from 11.0% to 12.0% (+9%) benchmark · Aug 6
Amazon
- Nova 2.0 Pro Preview Humanity's Last Exam improved from 8.9% to 9.4% (+6%) benchmark · Aug 6
Meta AI
- Muse Spark 1.2 added to the tracker: $1.25/$4.25 per 1M, 1M context. tracked · Aug 5
Poolside
- Laguna XS 2.1 vendor list price raised from $0.12 to $0.2 per 1M output tokens (+67%) list price · Aug 1
- Laguna XS 2.1 vendor list price raised from $0.06 to $0.1 per 1M input tokens (+67%) list price · Aug 1