FrontierCode 2.0 leaderboard
- Claude Fable 5.2 leads
- 49.1%
- models measured
- 55
- lead over #2
- +2.7 pts
- median
- 21.6%
Leader on FrontierCode 2.0 · read November 14
# FrontierCode 2.0, read November 14
leaders = {
"Claude Fable 5.2": 49.1%, "Claude Opus 5.5": 46.4%, "GPT-6 Astra": 42.4%,
}
On November 14, 2026, 55 of our catalog's models had a FrontierCode 2.0 result, and Anthropic Claude Fable 5.2 led with 49.1%, ahead of Anthropic Claude Opus 5.5 at 46.4%.
Software engineering tasks, version 2.0 of FrontierCode. Share solved. Run by Artificial Analysis on its own harness.
Role in the index
A coding board we show beside the Vibe Coding Index, not inside it: the index keeps the six boards it was defined on, so a score does not move every time a board is added. Read the methodology
| # | Model | Result | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | Claude Fable 5.2Anthropic | 49.1% | 87.8 | $40.00 | |
| 2 | Claude Opus 5.5Anthropic | 46.4% | 87.3 | $20.00 | |
| 3 | GPT-6 AstraOpenAI | 42.4% | 83.8 | $50.00 | |
| 4 | GPT-6.1 TerraOpenAI | 42.3% | 76.4 | $5.00 | |
| 5 | Claude Sonnet 5.5Anthropic | 41.7% | 84.3 | $10.00 | |
| 6 | GPT-6.1 SolOpenAI | 40.6% | 79.8 | $10.00 | |
| 7 | Gemini 4 CarbonGoogle | 40.6% | 86.6 | $10.00 | |
| 8 | Gemini 4 HydrogenGoogle | 39.0% | 77.9 | $7.50 | |
| 9 | Claude Fable 5.1Anthropic | 35.7% | 82.8 | $50.00 | |
| 10 | GPT-6 SolOpenAI | 35.6% | 72.1 | $10.00 | |
| 11 | Claude Haiku 5.5Anthropic | 34.9% | 68.4 | $0.50 | |
| 12 | Hy4Tencent | 31.4% | 65.2 | $1.2505 | |
| 13 | Claude Opus 5Anthropic | 31.2% | 81.7 | $25.00 | |
| 14 | Grok 4.7xAI | 28.7% | 64.8 | $6.00 | |
| 15 | DeepSeek V4.1 Flash 1017DeepSeek | 28.5% | 66.9 | $0.60 | |
| 16 | Muse WonderMeta | 27.9% | 71.8 | $20.00 | |
| 17 | Muse EmberMeta | 27.2% | 64.9 | $0.50 | |
| 18 | Gemini 4 ArgonGoogle | 25.7% | 80.3 | $10.00 | |
| 19 | Kimi K3.1 ProMoonshot AI | 25.5% | 79.0 | $10.00 | |
| 20 | T Inkling 2 SmallThinking Machines | 25.4% | 48.9 | $1.00 | |
| 21 | GPT-6 LunaOpenAI | 25.2% | 54.9 | $0.50 | |
| 22 | MiMo-V2.6-ProXiaomi | 24.0% | 61.8 | $0.87 | |
| 23 | GPT-5.6 TerraOpenAI | 23.9% | 63.5 | $12.00 | |
| 24 | Qwen3.8 Max (0902)Qwen | 23.4% | 67.3 | $6.00 | |
| 25 | Muse Spark 1.3Meta | 23.1% | 56.1 | $4.25 | |
| 26 | MiniMax M3MiniMax | 21.9% | 33.6 | $1.20 | |
| 27 | MiMo-V2.6-FlashXiaomi | 21.6% | 54.8 | $0.28 | |
| 28 | GLM 5.3Z.ai | 21.6% | 68.5 | $4.40 | |
| 29 | MiniMax M3.1 FlashMiniMax | 20.8% | 55.7 | $0.60 | |
| 30 | Grok 4.6xAI | 19.8% | 63.9 | $6.00 | |
| 31 | Gemini 3.8 FlashGoogle | 18.7% | 60.3 | $3.75 | |
| 32 | Claude Sonnet 5Anthropic | 17.8% | 54.6 | $10.00 | |
| 33 | GLM 5.3 FlashZ.ai | 17.6% | 56.2 | $0.50 | |
| 34 | GPT-5.6 LunaOpenAI | 17.6% | 55.3 | $1.20 | |
| 35 | DeepSeek V4 Pro 0813DeepSeek | 17.1% | 56.1 | $1.98 | |
| 36 | Ling 3.1 FlashinclusionAI | 17.0% | Not benchmarked | $0.135 | |
| 37 | Mistral Large 4Mistral AI | 15.0% | 60.9 | $2.10 | |
| 38 | Qwen3.8 27BQwen | 14.9% | 69.7 | $2.55 | |
| 39 | Step 5 PreviewStepFun | 12.0% | 57.4 | $2.70 | |
| 40 | Kimi K3Moonshot AI | 10.9% | 61.4 | $15.00 | |
| 41 | Muse Spark OpenMeta | 7.6% | 43.4 | $4.25 | |
| 42 | Nemotron 3.5 Super VLNVIDIA | 6.1% | 14.8 | $0.50 | |
| 43 | Gemini 3.5 Flash LiteGoogle | 6.0% | 48.9 | $2.50 | |
| 44 | T Inkling SmallThinking Machines | 4.8% | 27.3 | $1.20 | |
| 45 | Solar Pro 4Upstage | 4.1% | Not benchmarked | $0.60 | |
| 46 | T InklingThinking Machines | 3.9% | 29.2 | $4.05 | |
| 47 | Mistral Medium 3.5Mistral AI | 1.8% | 25.4 | $7.50 | |
| 48 | Nemotron 3 UltraNVIDIA | 1.1% | 30.0 | $2.30 | |
| 49 | Gemini 3.1 Flash LiteGoogle | 0.6% | Not benchmarked | $1.50 | |
| 50 | Command ACohere | 0.0% | Not benchmarked | $10.00 | |
| 51 | Gemma 4 31BGoogle | 0.0% | 21.9 | $0.40 | |
| 52 | Granite 4.2 8BIBM Granite | 0.0% | Not benchmarked | $0.20 | |
| 53 | Mistral Small 4Mistral AI | 0.0% | Not benchmarked | $0.60 | |
| 54 | Nemotron 3.5 LightningNVIDIA | 0.0% | 22.6 | $0.18 | |
| 55 | gpt-oss-120bOpenAI | 0.0% | Not benchmarked | $0.45 |
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.