LLM Leaderboard
Ranked by Intelligence Index · source AA
Artificial Analysis composite of reasoning, knowledge, coding, math.
#1 by Intelligence Index
Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)
59.9
Anthropic · in $10.0 / out $50.0 per 1M · 57 tok/s
| # | Model | Intelligence |
|---|---|---|
| 1 | Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) | 59.9 |
| 2 | GPT-5.6 Sol (max) | 58.9 |
| 3 | GPT-5.6 Sol (xhigh) | 57.7 |
| 4 | Kimi K3 | 57.1 |
| 5 | GPT-5.6 Sol (high) | 55.9 |
| 6 | Claude Opus 4.8 (Adaptive Reasoning, Max Effort) | 55.7 |
| 7 | GPT-5.6 Terra (max) | 55.0 |
| 8 | GPT-5.5 (xhigh) | 54.8 |
| 9 | Grok 4.5 (high) | 53.8 |
| 10 | GPT-5.6 Sol (medium) | 53.6 |
| 11 | Claude Opus 4.7 (Adaptive Reasoning, Max Effort) | 53.5 |
| 12 | Claude Sonnet 5 (Adaptive Reasoning, Max Effort) | 53.4 |
| 13 | GPT-5.5 (high) | 53.1 |
| 14 | GPT-5.6 Terra (xhigh) | 51.6 |
| 15 | GPT-5.4 (xhigh) | 51.4 |
| 16 | GPT-5.6 Luna (max) | 51.2 |
| 17 | GLM-5.2 (max) | 51.1 |
| 18 | Muse Spark 1.1 (xhigh) | 50.6 |
| 19 | GPT-5.5 (medium) | 50.4 |
| 20 | Gemini 3.5 Flash (high) | 50.2 |
| 21 | GPT-5.6 Sol (low) | 49.4 |
| 22 | GPT-5.6 Luna (xhigh) | 49.1 |
| 23 | GPT-5.6 Terra (high) | 49.0 |
| 24 | Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) | 47.2 |
| 25 | Gemini 3.1 Pro Preview | 46.5 |
| 26 | GPT-5.6 Luna (high) | 46.1 |
| 27 | Qwen3.7 Max | 46.0 |
| 28 | GPT-5.6 Terra (medium) | 45.6 |
| 29 | Gemini 3.5 Flash (medium) | 45.4 |
| 30 | MiniMax-M3 | 44.4 |
| 31 | DeepSeek V4 Pro (Reasoning, Max Effort) | 44.3 |
| 32 | GPT-5.3 Codex (xhigh) | 44.3 |
| 33 | Kimi K2.6 | 44.2 |
| 34 | Claude Opus 4.6 (Adaptive Reasoning, Max Effort) | 43.7 |
| 35 | GPT-5.5 (low) | 43.5 |
| 36 | Muse Spark | 43.1 |
| 37 | Claude Opus 4.7 (Non-reasoning, High Effort) | 42.7 |
| 38 | GPT-5.2 (xhigh) | 42.2 |
| 39 | MiMo-V2.5-Pro | 42.2 |
| 40 | Kimi K2.7 Code | 41.9 |
| 41 | Claude Sonnet 5 (Non-reasoning, High Effort) | 41.7 |
| 42 | GPT-5.6 Sol (Non-reasoning) | 41.2 |
| 43 | Hy3 | 41.2 |
| 44 | Nex-N2-Pro | 41.0 |
| 45 | DeepSeek V4 Pro (Reasoning, High Effort) | 40.8 |
| 46 | Claude Opus 4.5 (Reasoning) | 40.8 |
| 47 | Inkling (xhigh) | 40.7 |
| 48 | GPT-5.6 Terra (low) | 40.5 |
| 49 | MiMo-V2-Pro | 40.3 |
| 50 | DeepSeek V4 Flash (Reasoning, Max Effort) | 40.3 |
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.