CorpFin (Vals AI) leaderboard
- Claude Opus 5 leads
- 73.2%
- models measured
- 86
- lead over #2
- +1.4 pts
- median
- 61.9%
Leader on CorpFin (Vals AI) · read October 22
On October 22, 2026, 91 of our catalog's models had a CorpFin (Vals AI) result, and Anthropic Claude Opus 5 led with 73.2%, ahead of Anthropic Claude Fable 5 at 71.8%.
Corporate finance questions answered from long credit agreements. Share answered correctly, as run by Vals AI. An archived board.
Role in the index
CorpFin (Vals AI) is shown as published and does not feed any of our indexes. Read the methodology
| # | Model | Result | Latency, as printed | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|---|
| 1 | Claude Opus 5Anthropic | 73.2% | 29 s | 83.0 | $25.00 | |
| 2 | Claude Fable 5Anthropic | 71.8% | 53 s | 79.3 | $50.00 | |
| 3 | Kimi K3Moonshot AI | 71.6% | 38 s | 62.3 | $15.00 | |
| 4 | Muse Spark 1.1Meta | 71.3% | 28 s | 70.3 | $4.25 | |
| 5 | Muse Spark 1.2Meta | 70.9% | 30 s | 55.7 | $4.25 | |
| 6 | T Inkling SmallThinking Machines | 69.6% | 45 s | 27.3 | $1.20 | |
| 7 | T InklingThinking Machines | 68.6% | 29 s | 31.5 | $4.05 | |
| 8 | Grok 4.3xAI | 68.5% | 36 s | 37.4 | $2.50 | |
| 9 | GPT-5.5OpenAI | 68.4% | 23 s | 71.3 | $30.00 | |
| 10 | Kimi K2.5Moonshot AI | 68.3% | 1m 21s | 38.6 | $2.85 | |
| 11 | MiniMax M3MiniMax | 68.1% | 24 s | 39.2 | $1.20 | |
| 12 | Grok 4.5xAI | 67.4% | 1m 32s | 58.6 | $6.00 | |
| 13 | Claude Opus 4.6Anthropic | 67.0% | 21 s | 57.2 | $25.00 | |
| 14 | Claude Sonnet 5Anthropic | 67.0% | 19 s | 57.7 | $10.00 | |
| 15 | Kimi K2.6Moonshot AI | 66.7% | 1m 20s | 52.1 | $4.00 | |
| 16 | Claude Opus 4.8Anthropic | 66.7% | 20 s | 65.0 | $25.00 | |
| 17 | Gemini 3 Flash PreviewGoogle | 66.4% | 11 s | 43.7 | $3.00 | |
| 18 | Grok 4.6xAI | 66.2% | 24 s | 65.7 | $6.00 | |
| 19 | GLM 5.2Z.ai | 66.1% | 25 s | 61.1 | $4.40 | |
| 20 | Claude Opus 4.7Anthropic | 66.1% | 18 s | 67.5 | $25.00 | |
| 21 | GPT-5.2OpenAI | 65.9% | 26 s | 55.9 | $14.00 | |
| 22 | Qwen3.8 Max (0902)Qwen | 65.8% | 47 s | 62.4 | $6.00 | |
| 23 | Nemotron 3 UltraNVIDIA | 65.5% | 15 s | 30.0 | $2.40 | |
| 24 | DeepSeek V4 Pro 0813DeepSeek | 65.4% | 33 s | 56.1 | $1.98 | |
| 25 | Claude Sonnet 4.6Anthropic | 65.3% | 18 s | 57.3 | $15.00 | |
| 26 | GPT-5.6 TerraOpenAI | 65.3% | 7.7 s | 63.9 | $12.00 | |
| 27 | Qwen3.5 Plus 2026-04-20Qwen | 65.3% | 2m 36s | 39.5 | $1.80 | |
| 28 | GPT-5.4OpenAI | 65.3% | 1m 22s | 64.4 | $15.00 | |
| 29 | Claude Opus 4.5Anthropic | 65.1% | 18 s | 44.0 | $25.00 | |
| 30 | Gemini 3.5 FlashGoogle | 64.7% | 8.7 s | 39.8 | $9.00 | |
| 31 | Gemini 3.1 Pro PreviewGoogle | 64.5% | 25 s | 32.3 | $12.00 | |
| 32 | GLM 5.1Z.ai | 64.5% | 1m 08s | 49.2 | $4.40 | |
| 33 | GPT-5.6 SolOpenAI | 64.4% | 16 s | 73.5 | $20.00 | |
| 34 | GPT-5.6 LunaOpenAI | 64.2% | 17 s | 58.0 | $1.20 | |
| 35 | GPT-5.1OpenAI | 63.8% | 35 s | 40.4 | $10.00 | |
| 36 | Qwen3.7 MaxQwen | 63.7% | 20 s | 54.1 | $4.425 | |
| 37 | Grok 4.20xAI | 63.7% | 9.7 s | 33.3 | $2.50 | |
| 38 | Qwen3.5-FlashQwen | 63.6% | 35 s | 41.1 | $0.26 | |
| 39 | Gemini 3.6 FlashGoogle | 63.3% | 6.3 s | 62.5 | $3.75 | |
| 40 | GPT-4.1OpenAI | 63.0% | 10 s | Not benchmarked | $8.00 | |
| 41 | GLM 5Z.ai | 62.9% | 1m 09s | 41.7 | $3.20 | |
| 42 | Qwen3.6 27BQwen | 62.3% | 1m 03s | 36.8 | $2.70 | |
| 43 | Claude Sonnet 4.5Anthropic | 62.0% | 21 s | 39.5 | $15.00 | |
| 44 | Ling 3.0 FlashinclusionAI | 61.9% | 12 s | 29.6 | $0.1215 | |
| 45 | Qwen3.6 PlusQwen | 61.9% | 36 s | 44.1 | $1.95 | |
| 46 | DeepSeek V4 Flash 0731DeepSeek | 61.9% | 16 s | 51.8 | $0.396 | |
| 47 | MiMo-V2.5-ProXiaomi | 61.4% | 19 s | 31.6 | $0.9152 | |
| 48 | DeepSeek V4 Pro 0423DeepSeek | 61.4% | 1m 19s | 38.7 | $2.976 | |
| 49 | MiniMax M2.7MiniMax | 61.2% | 11 s | 36.2 | $1.20 | |
| 50 | GPT-5.4 NanoOpenAI | 61.2% | 9.4 s | 44.8 | $1.25 | |
| 51 | GPT-5OpenAI | 61.1% | 2m 41s | 38.8 | $10.00 | |
| 52 | Mistral Large 3 2512Mistral AI | 61.0% | 49 s | Not benchmarked | $1.50 | |
| 53 | GLM 4.5Z.ai | 61.0% | 1m 35s | Not benchmarked | $2.20 | |
| 54 | GPT-5.4 MiniOpenAI | 60.9% | 52 s | 35.8 | $4.50 | |
| 55 | Gemini 2.5 ProGoogle | 60.8% | 27 s | 19.1 | $10.00 | |
| 56 | Gemini 3.5 Flash LiteGoogle | 60.7% | 3.8 s | 49.0 | $2.50 | |
| 57 | Claude Haiku 4.5Anthropic | 60.6% | 14 s | 35.6 | $5.00 | |
| 58 | Kimi K2 ThinkingMoonshot AI | 60.6% | 1m 33s | Not benchmarked | $2.50 | |
| 59 | GPT-5 MiniOpenAI | 60.2% | 1m 14s | 34.2 | $2.00 | |
| 60 | MiMo-V2.5Xiaomi | 59.9% | 18 s | 50.0 | $0.336 | |
| 61 | o3OpenAI | 59.7% | 19 s | Not benchmarked | $8.00 | |
| 62 | Grok 3RetiredxAI | 59.7% | 24 s | Not benchmarked | $15.00 | |
| 63 | MiniMax M2.5MiniMax | 59.6% | 17 s | 37.2 | $1.20 | |
| 64 | Gemini 3.1 Flash Lite PreviewGoogle | 59.4% | 4.0 s | 22.7 | $1.50 | |
| 65 | o4 MiniOpenAI | 59.0% | 18 s | Not benchmarked | $4.40 | |
| 66 | MiniMax M2.1MiniMax | 58.9% | 22 s | Not benchmarked | $1.20 | |
| 67 | Mistral Medium 3.5Mistral AI | 58.8% | 44 s | 25.5 | $7.50 | |
| 68 | gpt-oss-120bOpenAI | 58.2% | 43 s | Not benchmarked | $0.50 | |
| 69 | GPT-4.1 MiniOpenAI | 57.9% | 6.4 s | Not benchmarked | $1.60 | |
| 70 | Gemini 2.5 Flash LiteGoogle | 57.6% | 54 s | Not benchmarked | $0.40 | |
| 71 | GLM 4.6Z.ai | 56.8% | 1m 18s | 8.8 | $2.20 | |
| 72 | DeepSeek V3 0324DeepSeek | 54.7% | 37 s | Not benchmarked | $1.07 | |
| 73 | Trinity Large ThinkingArcee AI | 54.7% | 12 s | Not benchmarked | $0.80 | |
| 74 | Nemotron 3.5 LightningNVIDIA | 54.2% | 39 s | 22.6 | $0.147 | |
| 75 | R1RetiredDeepSeek | 54.1% | 47 s | Not benchmarked | $2.50 | |
| 76 | gpt-oss-20bOpenAI | 53.1% | 34 s | Not benchmarked | $0.15 | |
| 77 | DeepSeek V3.2DeepSeek | 51.0% | 1m 39s | 30.0 | $0.88 | |
| 78 | Mistral Medium 3.1Mistral AI | 50.7% | 30 s | Not benchmarked | $2.00 | |
| 79 | Kimi K2 0711Moonshot AI | 50.4% | 41 s | Not benchmarked | $2.30 | |
| 80 | Llama 4 MaverickMeta Llama | 49.7% | 4.7 s | Not benchmarked | $1.00 | |
| 81 | Llama 4 ScoutMeta Llama | 46.8% | 9.7 s | Not benchmarked | $0.50 | |
| 82 | GLM 4.7Z.ai | 46.4% | 1m 06s | 55.1 | $2.20 | |
| 83 | Command ACohere | 46.0% | 14 s | Not benchmarked | $10.00 | |
| 84 | GPT-4o (2024-11-20)OpenAI | 45.9% | 6.5 s | Not benchmarked | $10.00 | |
| 85 | GPT-4o-miniOpenAI | 45.5% | 0.0 s | Not benchmarked | $0.60 | |
| 86 | o3 MiniOpenAI | 45.3% | 31 s | Not benchmarked | $4.40 | |
| 87 | GPT-4.1 NanoOpenAI | 42.1% | 5.0 s | Not benchmarked | $0.40 | |
| 88 | Gemini 1.5 ProRetiredGoogle | 40.5% | 38 s | Not benchmarked | $5.00 | |
| 89 | GPT-4o (2024-08-06)OpenAI | 39.4% | 0.1 s | Not benchmarked | $10.00 | |
| 90 | Gemini 1.5 FlashRetiredGoogle | 38.2% | 28 s | Not benchmarked | $0.30 | |
| 91 | Gemini 2.0 FlashRetiredGoogle | 33.7% | 30 s | Not benchmarked | $0.40 |
Only models in our catalog are listed, and each appears once with its best published run. 137 rows were read from the source on October 22, 2026, and rows that match no catalog model are left out. Every other column is a figure the board publishes for the same run.