HLE-Diamond (Artificial Analysis run) leaderboard
- Claude Fable 5.2 leads
- 80.2%
- models measured
- 74
- lead over #2
- +11.1 pts
- median
- 18.1%
Leader on HLE-Diamond (Artificial Analysis run) · read November 14
On November 14, 2026, 77 of our catalog's models had a HLE-Diamond (Artificial Analysis run) result, and Anthropic Claude Fable 5.2 led with 80.2%, ahead of Google Gemini 4 Carbon at 69.1%.
A smaller, hand-checked subset of Humanity's Last Exam. Share answered correctly. Run by Artificial Analysis on its own harness.
Role in the index
HLE-Diamond (Artificial Analysis run) is shown as published and does not feed any of our indexes. Read the methodology
| # | Model | Result | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | Claude Fable 5.2Anthropic | 80.2% | 87.8 | $40.00 | |
| 2 | Gemini 4 CarbonGoogle | 69.1% | 86.6 | $10.00 | |
| 3 | Claude Opus 5.5Anthropic | 67.9% | 87.3 | $20.00 | |
| 4 | GPT-6 AstraOpenAI | 63.8% | 83.8 | $50.00 | |
| 5 | Claude Fable 5.1Anthropic | 60.1% | 82.8 | $50.00 | |
| 6 | Gemini 4 HydrogenGoogle | 58.8% | 77.9 | $7.50 | |
| 7 | GPT-6.1 SolOpenAI | 58.8% | 79.8 | $10.00 | |
| 8 | Gemini 4 ArgonGoogle | 56.7% | 80.3 | $10.00 | |
| 9 | Claude Sonnet 5.5Anthropic | 53.3% | 84.3 | $10.00 | |
| 10 | GPT-6.1 TerraOpenAI | 47.3% | 76.4 | $5.00 | |
| 11 | GPT-6 SolOpenAI | 45.2% | 72.1 | $10.00 | |
| 12 | Claude Opus 5Anthropic | 44.8% | 81.7 | $25.00 | |
| 13 | Muse WonderMeta | 39.0% | 71.8 | $20.00 | |
| 14 | Kimi K3.1 ProMoonshot AI | 37.3% | 79.0 | $10.00 | |
| 15 | Muse Spark 1.3Meta | 33.7% | 56.1 | $4.25 | |
| 16 | Gemini 3.8 FlashGoogle | 33.1% | 60.3 | $3.75 | |
| 17 | Hy4Tencent | 32.7% | 65.2 | $1.2505 | |
| 18 | Gemini 3.1 Pro PreviewGoogle | 31.4% | 32.3 | $12.00 | |
| 19 | Claude Haiku 5.5Anthropic | 29.6% | 68.4 | $0.50 | |
| 20 | GPT-5.6 TerraOpenAI | 28.8% | 63.5 | $12.00 | |
| 21 | T Inkling 2 SmallThinking Machines | 28.7% | 48.9 | $1.00 | |
| 22 | MiniMax M3.1 FlashMiniMax | 28.1% | 55.7 | $0.60 | |
| 23 | Kimi K3Moonshot AI | 26.9% | 61.4 | $15.00 | |
| 24 | Muse Spark OpenMeta | 26.3% | 43.4 | $4.25 | |
| 25 | DeepSeek V4.1 Flash 1017DeepSeek | 25.3% | 66.9 | $0.60 | |
| 26 | Muse EmberMeta | 25.1% | 64.9 | $0.50 | |
| 27 | Step 5 PreviewStepFun | 24.3% | 57.4 | $2.70 | |
| 28 | Qwen3.8 Max (0902)Qwen | 24.0% | 67.3 | $6.00 | |
| 29 | Grok 4.7xAI | 23.6% | 64.8 | $6.00 | |
| 30 | GPT-6 LunaOpenAI | 22.2% | 54.9 | $0.50 | |
| 31 | Grok 4.6xAI | 21.5% | 63.9 | $6.00 | |
| 32 | DeepSeek V4 Flash Vision ExpDeepSeek | 21.1% | Not benchmarked | $0.9834 | |
| 33 | GPT-5.6 LunaOpenAI | 20.9% | 55.3 | $1.20 | |
| 34 | DeepSeek V4 Pro 0813DeepSeek | 20.5% | 56.1 | $1.98 | |
| 35 | Qwen3.8 2.4T A95BQwen | 20.1% | Not benchmarked | $6.00 | |
| 36 | DeepSeek V4.1 FlashDeepSeek | 19.1% | 56.6 | $0.79 | |
| 37 | GLM 5.3 FlashZ.ai | 18.2% | 56.2 | $0.50 | |
| 38 | GLM 5.3Z.ai | 17.9% | 68.5 | $4.40 | |
| 39 | GLM 5.2Z.ai | 17.7% | 61.1 | $4.40 | |
| 40 | Mistral Large 4Mistral AI | 17.3% | 60.9 | $2.10 | |
| 41 | Ling 3.1 FlashinclusionAI | 17.2% | Not benchmarked | $0.135 | |
| 42 | MiniMax M3MiniMax | 16.0% | 33.6 | $1.20 | |
| 43 | Qwen3.8 27BQwen | 15.3% | 69.7 | $2.55 | |
| 44 | MiMo-V2.6-ProXiaomi | 15.2% | 61.8 | $0.87 | |
| 45 | Kimi K2.7 CodeMoonshot AI | 14.8% | 53.4 | $4.00 | |
| 46 | Qwen3.7 PlusQwen | 13.9% | 33.2 | $1.28 | |
| 47 | MiMo-V2.6-FlashXiaomi | 13.7% | 54.8 | $0.28 | |
| 48 | Qwen3.5 397B A17BQwen | 12.4% | Not benchmarked | $2.34 | |
| 49 | T Inkling SmallThinking Machines | 12.1% | 27.3 | $1.20 | |
| 50 | Hy3Tencent | 11.6% | Not benchmarked | $0.528 | |
| 51 | LongCat 2.0Meituan | 11.5% | Not benchmarked | $1.20 | |
| 52 | T InklingThinking Machines | 10.6% | 29.2 | $4.05 | |
| 53 | Gemini 3.5 Flash LiteGoogle | 9.9% | 48.9 | $2.50 | |
| 54 | Gemini 3.1 Flash LiteGoogle | 9.0% | Not benchmarked | $1.50 | |
| 55 | Nemotron 3.5 Super VLNVIDIA | 8.8% | 14.8 | $0.50 | |
| 56 | Nemotron 3 UltraNVIDIA | 8.5% | 30.0 | $2.30 | |
| 57 | Ling 3.0 Flash VLinclusionAI | 7.6% | Not benchmarked | $0.18 | |
| 58 | Qwen3.5-122B-A10BQwen | 7.3% | Not benchmarked | $2.08 | |
| 59 | Gemma 4 31BGoogle | 7.2% | 21.9 | $0.40 | |
| 60 | Ling 3.0 FlashinclusionAI | 7.2% | 29.6 | $0.1215 | |
| 61 | Qwen3.6 35B A3BQwen | 6.9% | Not benchmarked | $0.925 | |
| 62 | Gemma 4 26B A4BGoogle | 6.2% | Not benchmarked | $0.60 | |
| 63 | Trinity Large ThinkingRetiredArcee AI | 6.0% | Not benchmarked | $0.80 | |
| 64 | Ling 3.0 Flash FinRetiredinclusionAI | 5.4% | Not benchmarked | $0.1232 | |
| 65 | Command ACohere | 5.1% | Not benchmarked | $10.00 | |
| 66 | Mercury 2.5RetiredInception | 5.1% | 1.6 | $0.15 | |
| 67 | Nemotron 3 SuperNVIDIA | 4.9% | Not benchmarked | $0.425 | |
| 68 | gpt-oss-20bOpenAI | 4.8% | Not benchmarked | $0.14 | |
| 69 | Qwen3.5-9BQwen | 4.8% | Not benchmarked | $0.15 | |
| 70 | Mistral Medium 3.5Mistral AI | 4.7% | 25.4 | $7.50 | |
| 71 | Solar Pro 4Upstage | 4.7% | Not benchmarked | $0.60 | |
| 72 | Nemotron 3.5 LightningNVIDIA | 4.5% | 22.6 | $0.18 | |
| 73 | Mistral Small 4Mistral AI | 4.4% | Not benchmarked | $0.60 | |
| 74 | Qwen3 Coder NextQwen | 4.3% | Not benchmarked | $0.80 | |
| 75 | Granite 4.2 8BIBM Granite | 4.0% | Not benchmarked | $0.20 | |
| 76 | gpt-oss-120bOpenAI | 3.8% | Not benchmarked | $0.45 | |
| 77 | Nemotron 3 Nano 30B A3BNVIDIA | 3.5% | Not benchmarked | $0.24 |
Only models in our catalog are listed, and each appears once with its best published run. 214 rows were read from the source on November 14, 2026, and rows that match no catalog model are left out.