LLM Leaderboard
Ranked by Intelligence Index · source AA
Artificial Analysis composite of reasoning, knowledge, coding, math.
#1 by Intelligence Index
Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)
53.4
Anthropic · in $10.0 / out $50.0 per 1M · 70 tok/s
| # | Model | Intelligence |
|---|---|---|
| 1 | Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) | 53.4 |
| 2 | Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback) | 53.2 |
| 3 | Muse Spark 1.3 (max) | 53.0 |
| 4 | GPT-6 Astra (max) | 52.8 |
| 5 | GPT-6 Astra (xhigh) | 52.5 |
| 6 | Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback) | 51.2 |
| 7 | GPT-6 Astra (high) | 51.0 |
| 8 | Claude Opus 5 (Adaptive Reasoning, Max Effort) | 50.7 |
| 9 | Claude Opus 5 (Adaptive Reasoning, Xhigh Effort) | 49.7 |
| 10 | Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) | 49.7 |
| 11 | GPT-6 Astra (medium) | 49.7 |
| 12 | Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback) | 49.1 |
| 13 | Claude Opus 5 (Adaptive Reasoning, High Effort) | 48.2 |
| 14 | Muse Spark 1.3 (max) | 48.2 |
| 15 | GPT-5.6 Sol (max) | 47.1 |
| 16 | Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback) | 47.0 |
| 17 | GPT-6 Astra (low) | 46.0 |
| 18 | Qwen3.8 Max (0902) | 45.4 |
| 19 | GPT-6 Astra (Non-reasoning) | 45.2 |
| 20 | Muse Spark 1.3 (xhigh) | 45.2 |
| 21 | Claude Opus 5 (Adaptive Reasoning, Medium Effort) | 45.1 |
| 22 | GLM-5.3 (max) | 44.9 |
| 23 | Grok 4.6 (high) | 44.4 |
| 24 | Grok 4.6 (xhigh) | 44.3 |
| 25 | GPT-5.6 Sol (xhigh) | 44.1 |
| 26 | Kimi K3 (max) | 43.8 |
| 27 | Grok 4.6 (medium) | 43.0 |
| 28 | GPT-5.6 Sol (high) | 42.5 |
| 29 | GPT-5.6 Terra (max) | 42.3 |
| 30 | Claude Opus 4.8 (Adaptive Reasoning, Max Effort) | 42.0 |
| 31 | GLM 5.3 Flash | 41.9 |
| 32 | Gemini 3.8 Flash (high) | 41.2 |
| 33 | Claude Opus 4.7 (Adaptive Reasoning, Max Effort) | 40.7 |
| 34 | Qwen3.8 Max | 40.3 |
| 35 | Gemini 3.8 Flash (medium) | 40.0 |
| 36 | Qwen3.8 2.4T A95B | 40.0 |
| 37 | Qwen3.8-Flash-Next | 39.9 |
| 38 | Muse Spark 1.2 (xhigh) | 39.8 |
| 39 | Claude Opus 5 (Adaptive Reasoning, Low Effort) | 39.8 |
| 40 | Gemini 3.7 Flash (medium) | 39.6 |
| 41 | DeepSeek V4.1 Flash (Reasoning, Max Effort) | 39.5 |
| 42 | GPT-5.6 Sol (medium) | 39.5 |
| 43 | Gemini 3.7 Flash (high) | 39.4 |
| 44 | Grok 4.5 (high) | 39.1 |
| 45 | GPT-5.4 (xhigh) | 39.0 |
| 46 | GPT-5.5 (xhigh) | 38.6 |
| 47 | Claude Sonnet 5 (Adaptive Reasoning, Max Effort) | 38.4 |
| 48 | GPT-5.6 Terra (xhigh) | 38.2 |
| 49 | GPT-5.6 Luna (max) | 37.5 |
| 50 | GPT-5.5 (high) | 37.3 |
Quality, pricing, and speed benchmarks via Artificial Analysis. Human preference Elo via LMArena. Refreshed daily at 06:00 UTC · last sync .
About the metrics
Intelligence Index
AA
Artificial Analysis composite of reasoning, knowledge, coding, math.
LMArena Elo
LMArena
Human preference Elo from chatbot arena head-to-head votes.
Coding Index
AA
AA composite — LiveCodeBench, SciCode, etc.
Math Index
AA
AA composite — MATH-500, AIME, etc.
MMLU-Pro
AA
Multi-domain knowledge & reasoning. Harder than MMLU.
GPQA Diamond
AA
Graduate-level science Q&A. Strong reasoning signal.
LiveCodeBench
AA
Competitive programming, contamination-resistant.
MATH-500
AA
Subset of the MATH benchmark.
AIME
AA
American Invitational Mathematics Examination — olympiad math.
Humanity's Last Exam
AA
Frontier reasoning benchmark; most models still score low.
SciCode
AA
Scientific coding tasks.