Best LLM
What is the best LLM for finance tasks?
Gemini 4 Carbon, Claude Fable 5.2, Claude Fable 5.1 and 3 others are practically level at the top in finance tasks.
Among the ten best, Gemini 4 Argon is the cheapest at $4 per million tokens.
The fastest among them is Claude Sonnet 5.5, at 103 tokens per second.
- Ranked models
- 113
- Tests
- 9
- Data as of
- Nov 10, 2026
Ranking: Finance
Place and value in the area, plus overall index, price and speed. Higher values are better.
| Rank | Model | Value | Overall index | Price (3:1) | Speed |
|---|---|---|---|---|---|
| 1 | Gemini 4 CarbonTop group | 123.8 | 127.0 | n/a | n/a |
| 2 | Claude Fable 5.2Top group | 123.5 | 130.9 | $16 | n/a |
| 3 | Claude Fable 5.1Top group | 121.8 | 124.1 | $20 | 59 |
| 4 | Gemini 4 ArgonTop group | 121.6 | 121.4 | $4 | 46 |
| 5 | Claude Opus 5.5Top group | 120.7 | 125.6 | $8 | 75 |
| 6 | Claude Sonnet 5.5Upper group | 119.4 | 121.8 | $4 | 103 |
| 7 | Claude Opus 5Upper group | 119.2 | 117.4 | $10 | 71 |
| 8 | Muse Wonder FullUpper group | 118.1 | 118.1 | n/a | n/a |
| 9 | Inkling 2 SmallMiddle group | 112.9 | n/a | n/a | n/a |
| 10 | Claude Fable 5Middle group | 112.7 | 112.8 | $20 | 50 |
| 11 | Muse Ember 1.0Middle group | 112.0 | 110.4 | $0.20 | n/a |
| 12 | Muse Spark 1.3Middle group | 112.0 | 107.3 | $2 | 102 |
| 13 | Muse WonderMiddle group | 112.0 | 113.3 | $8 | n/a |
| 14 | Gemini 3.8 FlashMiddle group | 112.0 | 106.8 | $1.50 | 81 |
| 15 | DeepSeek V4.1 Flash 1017Middle group | 108.0 | 106.7 | $0.26 | 126 |
| 16 | GLM-5.3Middle group | 107.2 | 100.6 | $2.15 | 81 |
| 17 | Hy4Middle group | 106.9 | n/a | n/a | n/a |
| 18 | GPT-6 AstraMiddle group | 106.5 | 120.2 | $20 | 58 |
| 19 | GPT-6.1 SolMiddle group | 105.9 | 118.0 | $4 | 44 |
| 20 | Grok 4.7Middle group | 105.7 | 103.3 | $3 | 47 |
Show all 113 ranked models
| Rank | Model | Value | Overall index | Price (3:1) | Speed |
|---|---|---|---|---|---|
| 21 | MiMo V2.6 ProMiddle group | 105.5 | 100.4 | $0.54 | 32 |
| 22 | Kimi K3Middle group | 105.3 | 101.1 | $6 | 61 |
| 23 | MiMo V2.6 FlashMiddle group | 105.1 | 97.3 | $0.18 | 38 |
| 24 | Gemini 3.7 FlashMiddle group | 104.9 | 107.9 | n/a | 72 |
| 25 | Kimi K3.1 ProMiddle group | 104.9 | 112.6 | $4 | 56 |
| 26 | Claude Opus 4.8Middle group | 104.5 | 99.5 | $10 | 71 |
| 27 | Claude Haiku 5.5Middle group | 104.3 | 102.8 | $0.20 | 164 |
| 28 | Mistral Large 4Middle group | 104.2 | 100.8 | $1.05 | n/a |
| 29 | DeepSeek-V4.1-FlashMiddle group | 103.7 | 99.5 | $0.26 | 118 |
| 30 | GPT-5.6 SolMiddle group | 103.6 | 109.1 | $8 | 75 |
| 31 | Hy4 previewMiddle group | 103.2 | n/a | n/a | n/a |
| 32 | GPT-5.6 TerraMiddle group | 102.5 | 100.8 | $4.50 | 72 |
| 33 | Claude Sonnet 5Middle group | 102.1 | 92.1 | $4 | 77 |
| 34 | Ember-1Middle group | 102.0 | n/a | n/a | n/a |
| 35 | Grok 4.6Middle group | 101.0 | 102.1 | $3 | 58 |
| 36 | MiniMax M3.1 FlashMiddle group | 100.9 | 100.0 | $0.26 | n/a |
| 37 | GLM-5.3-FlashMiddle group | 100.4 | 94.3 | $0.24 | 101 |
| 38 | GPT-5.5Middle group | 100.0 | 101.5 | $11.25 | 87 |
| 39 | GPT-5.6 LunaMiddle group | 99.3 | 93.8 | $0.45 | 123 |
| 40 | Step 5 PreviewMiddle group | 99.2 | 97.1 | $1.43 | 65 |
| 41 | GPT-6 SolMiddle group | 99.0 | 109.1 | $4 | 82 |
| 42 | GPT-6.1 TerraMiddle group | 99.0 | 106.5 | $2 | 62 |
| 43 | Claude Opus 4.7Middle group | 98.7 | 95.1 | $10 | 57 |
| 44 | Qwen 3.8 Max 0902Middle group | 98.7 | 100.8 | $2.48 | 36 |
| 45 | Gemini 3.5 FlashMiddle group | 98.5 | 97.8 | n/a | 142 |
| 46 | GPT-6 LunaMiddle group | 98.2 | 95.8 | $0.20 | 77 |
| 47 | Muse Spark OpenMiddle group | 97.7 | 97.5 | n/a | n/a |
| 48 | Gemini 3.6 FlashMiddle group | 96.3 | 97.6 | $1.50 | 110 |
| 49 | Claude Sonnet 4.6Middle group | 96.0 | 86.7 | $6 | 38 |
| 50 | Qwen 3.8 27BMiddle group | 94.8 | 82.8 | $0.66 | 66 |
| 51 | GLM-5.2Middle group | 94.3 | 89.0 | $2.15 | 73 |
| 52 | DeepSeek V4 Pro 0813Middle group | 94.2 | n/a | n/a | n/a |
| 53 | DeepSeek V4 Flash 0731Middle group | 93.2 | n/a | n/a | n/a |
| 54 | Grok 4.5Middle group | 92.5 | 94.8 | $3 | 52 |
| 55 | Qwen 3.7 MaxMiddle group | 87.7 | 85.4 | $2.48 | 50 |
| 56 | DeepSeek-V4-ProMiddle group | 87.5 | 87.2 | $0.99 | 44 |
| 57 | MiniMax M3Middle group | 87.3 | 83.0 | $0.52 | 82 |
| 58 | Kimi K2.6Middle group | 86.5 | 84.9 | $1.71 | 32 |
| 59 | GPT-5.4Middle group | 86.1 | 86.5 | $5.63 | 58 |
| 60 | Mercury 3Lower group | 84.1 | n/a | $0.16 | n/a |
| 61 | GLM-5.1Lower group | 83.3 | 82.8 | $2.15 | 39 |
| 62 | GPT-5Lower group | 82.3 | 77.5 | $3.44 | 83 |
| 63 | Reflection BeamLower group | 81.4 | 82.4 | n/a | n/a |
| 64 | Gemini 3.1 Pro PreviewLower group | 80.7 | 91.4 | $4.50 | 96 |
| 65 | MiMo-V2.5-ProLower group | 78.9 | 73.0 | $0.54 | 59 |
| 66 | Claude Opus 4.5Lower group | 78.7 | 74.4 | $10 | 44 |
| 67 | Qwen 3.7 PlusLower group | 78.5 | 83.4 | $0.48 | 50 |
| 68 | o3Lower group | 76.7 | 74.2 | $3.50 | 73 |
| 69 | Gemini 3 Flash PreviewLower group | 76.0 | 81.0 | $1.13 | 101 |
| 70 | GPT-5.4 miniLower group | 75.3 | 75.1 | $1.69 | 134 |
| 71 | Gemini 3.5 Flash-LiteLower group | 75.1 | 73.7 | $0.85 | 145 |
| 72 | GLM-5Lower group | 73.1 | 72.7 | $1.55 | 41 |
| 73 | Kimi K2.5Lower group | 72.5 | 70.3 | $1.14 | 37 |
| 74 | InklingLower group | 71.6 | 72.1 | $1.76 | n/a |
| 75 | MiMo-V2.5Lower group | 71.3 | 69.1 | $0.18 | 55 |
| 76 | GPT-5 miniLower group | 70.8 | 65.8 | $0.69 | 105 |
| 77 | Nemotron 3 UltraLower group | 70.5 | 66.8 | $0.93 | 92 |
| 78 | Nemotron 3.5 Super VL 120B A12BLower group | 69.1 | 68.1 | $0.24 | n/a |
| 79 | Claude Sonnet 4.5Lower group | 68.3 | 65.9 | $6 | 38 |
| 80 | Inkling SmallLower group | 65.9 | 68.7 | $0.52 | 61 |
| 81 | GPT-5.4 nanoLower group | 65.2 | 64.0 | $0.46 | 98 |
| 82 | Gemini 2.5 ProLower group | 64.8 | 61.4 | $3.44 | 74 |
| 83 | o4-miniLower group | 63.5 | 63.5 | n/a | n/a |
| 84 | MiniMax M2.5Lower group | 60.9 | 61.5 | $0.52 | 49 |
| 85 | Grok 4.3Lower group | 59.7 | 71.6 | $1.56 | 66 |
| 86 | MiniMax M2.7Lower group | 59.1 | 65.1 | $0.37 | 45 |
| 87 | Kimi K2 ThinkingLower group | 58.8 | 53.2 | $1.08 | 43 |
| 88 | GLM-4.5Lower group | 57.6 | 48.8 | $1 | 28 |
| 89 | Ling 3.0 FlashLower group | 56.3 | 47.9 | $0.032 | 106 |
| 90 | MiniMax M2.1Lower group | 56.0 | 57.3 | $0.52 | n/a |
| 91 | GPT-4.1Lower group | 55.7 | 43.6 | $3.50 | 81 |
| 92 | GPT OSS 120BLower group | 48.6 | 43.4 | $0.20 | 92 |
| 93 | Grok 4.20 ReasoningLower group | 48.1 | 58.3 | $1.56 | 66 |
| 94 | GPT-4.1 miniLower group | 47.8 | 32.3 | $0.70 | 95 |
| 95 | GLM-4.6Lower group | 45.7 | 45.9 | $0.88 | 28 |
| 96 | DeepSeek-R1Lower group | 43.7 | 44.5 | $1.15 | n/a |
| 97 | Mistral Large 3Lower group | 41.5 | 32.7 | $0.75 | 86 |
| 98 | DeepSeek-V3 0324Lower group | 38.3 | 33.6 | $0.44 | 39 |
| 99 | Kimi K2 0711Lower group | 37.1 | n/a | $1 | 39 |
| 100 | GLM-4.7Lower group | 36.8 | 56.2 | $0.78 | 25 |
| 101 | Mistral Medium 3.5Lower group | 36.7 | 46.3 | $3 | 118 |
| 102 | Nemotron 3.5 LightningLower group | 33.4 | 31.1 | $0.085 | 180 |
| 103 | GPT OSS 20BLower group | 30.4 | 30.9 | $0.06 | 85 |
| 104 | Mercury 2.5Lower group | 30.1 | 45.3 | n/a | n/a |
| 105 | GPT-5 nanoLower group | 29.9 | 32.4 | $0.14 | 136 |
| 106 | o3-miniLower group | 27.0 | 41.9 | n/a | n/a |
| 107 | Laguna XS.2Lower group | 24.5 | n/a | n/a | n/a |
| 108 | DeepSeek-V3Lower group | 24.5 | 19.7 | $0.37 | 40 |
| 109 | Llama 4 ScoutLower group | 5.0 | 4.7 | $0.15 | 62 |
| 110 | Mistral Small 3.1Lower group | 1.3 | -3.0 | n/a | n/a |
| 111 | Command ALower group | 0.5 | 0.4 | $4.38 | 2 |
| 112 | GPT-4o miniLower group | -3.6 | -7.3 | $0.26 | 66 |
| 113 | GPT-4.1 nanoLower group | -14.8 | -15.8 | n/a | 110 |
Price in dollars per million tokens, mixed from 3 parts input and 1 part output. Speed in tokens per second.
Gemini 4 CarbonRank 1123.8
Claude Fable 5.2Rank 2123.5
Claude Fable 5.1Rank 3121.8
Gemini 4 ArgonRank 4121.6
Claude Opus 5.5Rank 5120.7
Claude Sonnet 5.5Rank 6119.4
Claude Opus 5Rank 7119.2
Muse Wonder FullRank 8118.1
Inkling 2 SmallRank 9112.9
Claude Fable 5Rank 10112.7
Muse Ember 1.0Rank 11112.0
Muse Spark 1.3Rank 12112.0
Muse WonderRank 13112.0
Gemini 3.8 FlashRank 14112.0
DeepSeek V4.1 Flash 1017Rank 15108.0
GLM-5.3Rank 16107.2
Hy4Rank 17106.9
GPT-6 AstraRank 18106.5
GPT-6.1 SolRank 19105.9
Grok 4.7Rank 20105.7
Which tests count here?
9 tests by independent evaluators feed this area. Each test counts with the result of the model in question.
TaxEval v2
82 models tested · Vals AI
CorpFin v2
79 models tested · Vals AI
Tax Agent Bench
76 models tested · Vals AI
Finance Agent (v2)
72 models tested · Vals AI
EMB
68 models tested · Vals AI
MortgageTax
53 models tested · Vals AI
EMB v1.1
39 models tested · Vals AI
Finance Agent v2.1
37 models tested · Vals AI
PRBench Finance
16 models tested · Scale AI
Popular comparisons among the best
Other tasks
All LLMs comparedHow the overall index worksCompare several models