Live data
Assembling the frontier
Ranking the latest models and provider endpoints.
Loading model dataLive data
Ranking the latest models and provider endpoints.
Loading model dataLive model ledger
Top overall
Claude Fable 5.1
QI 53.4
Best value
DeepSeek V4 Flash 0731
$0.05/M
Fastest
Celeris-1
2028 tok/s
Largest context
Llama 4 Scout
1.3M
| # | Model | Quality | Price/M | Speed | Context |
|---|---|---|---|---|---|
| 1 | Claude Fable 5.1Anthropic | 53.4 | $20.0 | 68 tok/s | 1.0M |
| 2 | GPT-6 AstraOpenAI | 52.8 | $20.0 | 68 tok/s | 1.0M |
| 3 | GPT-6 Astra (xhigh)OpenAI | 52.5 | $20.0 | 62 tok/s | 1.0M |
| 4 | GPT-6 Astra (high)OpenAI | 51 | $20.0 | 57 tok/s | 1.0M |
| 5 | Claude Opus 5Anthropic | 50.7 | $10.0 | 59 tok/s | 1.0M |
| 6 | Claude Fable 5Anthropic | 49.7 | $20.0 | 69 tok/s | 1.0M |
| 7 | GPT-6 Astra (medium)OpenAI | 49.7 | $20.0 | 54 tok/s | 1.0M |
| 8 | Muse Spark 1.3Meta | 48.2 | $2.0 | 386 tok/s | 1.0M |
| 9 | GPT-5.6 SolOpenAI | 47.1 | $8.0 | 129 tok/s | 1.0M |
| 10 | GPT-6 Astra (low)OpenAI | 46 | $20.0 | 57 tok/s | 1.0M |
| 11 | Muse Spark 1.3 (xhigh)Meta | 45.2 | $2.0 | 328 tok/s | 1.0M |
| 12 | GLM-5.3Z AI | 44.9 | $1.7 | 416 tok/s | 1.0M |
| 13 | Grok 4.6 (high)SpaceXAI | 44.4 | $3.0 | 73 tok/s | 500K |
| 14 | Grok 4.6 (xhigh)SpaceXAI | 44.3 | $3.0 | 70 tok/s | 500K |
| 15 | GPT-5.6 Sol (xhigh)OpenAI | 44.1 | $8.0 | 95 tok/s | 1.0M |
| 16 | Kimi K3Kimi | 43.8 | $5.1 | 298 tok/s | 1.0M |
| 17 | Grok 4.6 (medium)SpaceXAI | 43 | $3.0 | 67 tok/s | 500K |
| 18 | GPT-5.6 Sol (high)OpenAI | 42.5 | $8.0 | 90 tok/s | 1.0M |
| 19 | GPT-5.6 TerraOpenAI | 42.3 | $4.5 | 167 tok/s | 1.0M |
| 20 | GLM-5.3-FlashZ AI | 41.9 | $0.11 | 558 tok/s | 1.0M |
| 21 | Gemini 3.8 Flash (high)Google | 41.2 | $1.5 | 339 tok/s | 1.0M |
| 22 | Qwen3.8 MaxAlibaba | 40.3 | $3.0 | 42 tok/s | 1.0M |
| 23 | Qwen3.8 2.4T A95BAlibaba | 40 | $2.8 | 167 tok/s | 984K |
| 24 | Qwen3.8-Flash-NextAlibaba | 39.9 | $0.23 | 124 tok/s | 1.0M |
| 25 | Muse Spark 1.2 (xhigh)Meta | 39.8 | $2.0 | 228 tok/s | 1.0M |
| 26 | Gemini 3.7 Flash (medium)Google | 39.6 | $1.5 | 348 tok/s | 1.0M |
| 27 | GPT-5.6 Sol (medium)OpenAI | 39.5 | $8.0 | 63 tok/s | 1.0M |
| 28 | DeepSeek V4.1 FlashDeepSeek | 39.5 | $0.52 | 289 tok/s | 1.0M |
| 29 | Gemini 3.7 Flash (high)Google | 39.4 | $1.5 | 336 tok/s | 1.0M |
| 30 | Grok 4.5 (high)SpaceXAI | 39.1 | $3.0 | 61 tok/s | 500K |
Showing top 30 of 175 ranked models
View all in Explore →Decision routes
The overall score opens the conversation. These guides finish it with task-specific methodology and picks.
Ranking library
Focused rankings for the decisions engineers actually make.
Methodology
Every model is scored using the Artificial Analysis Intelligence Index — a composite of GPQA Diamond, AIME 2025, LiveCodeBench, MMLU-Pro, and other benchmarks, weighted into a single 0-100 quality score. Speed, price, and context window are tracked live across providers.
The overall ranking is a starting point. For production decisions, narrow by use case using the category pages above, then compare finalists head-to-head on Compare.
Field notes
Claude Fable 5.1 leads on overall quality right now, but the best model depends on your priorities. Coding, cost, speed, and context length all shift the answer. Use the category rankings above to find the right fit.
DeepSeek V4 Flash 0731 currently offers one of the best quality-to-cost ratios. Open-source models on providers like Groq or Together can be even cheaper at strong quality levels.
Start with overall quality index, then narrow by what matters for your workload: cost per million tokens, output speed, context window, or a specific capability like coding or tool use. Use our Compare tool to put finalists head to head.
Celeris-1 leads on output speed right now at 2028 tokens/second. Speed matters most for real-time applications and agentic workflows with many sequential steps.
Llama 4 Scout has the biggest context window in this ranking at 1.3M. For a dedicated long-context comparison, see our largest context window page.
Data is pulled from Artificial Analysis and refreshed automatically. New models appear as soon as they have benchmark scores and provider endpoints. The ranking reflects the live state of the leaderboard.