An LLM leaderboard with rankings of model performance on legal-related queries and prompts submitted and scored by legal professionals. Compare LLM model performance, confidence intervals, pricing, and capabilities across top legal AI software—helping you determine which AI is best for legal research, contract review, workflow automation and other legal tasks.
| Rank | Model | Score | Rank spread | Price $/M | Context |
|---|---|---|---|---|---|
| 1 | Gemini 3.8 Flash· Google | 1094±22 | 1 - 8 | $1.50 / $7.50 input / output | 1M |
| 2 | Gemini 3.7 Flash· Google | 1088±16 | 1 - 8 | $1.50 / $7.50 input / output | 1M |
| 3 | Gemini 3.6 Flash· Google | 1078±10 | 1 - 9 | $1.50 / $7.50 input / output | 1M |
| 4 | GPT-6 Sol· OpenAI | 1078±33 | 1 - 26 | $4.00 / $15.00 input / output | 1.1M |
| 5 | GPT-6 Astra· OpenAI | 1077±24 | 1 - 18 | $10.00 / $50.00 input / output | 272K |
| 6 | Claude Fable 5.1· Anthropic | 1072±11 | 1 - 13 | $10.00 / $50.00 input / output | 1M |
| 7 | Gemini 3.5 Flash· Google | 1068±9 | 1 - 13 | $1.50 / $9.00 input / output | 1M |
| 8 | Gemini 3.1 Pro· Google | 1064±6 | 3 - 13 | $2.00 / $12.00 input / output | 1M |
| 9 | GPT-5.5· OpenAI | 1060±6 | 4 - 17 | $10.00 / $45.00 input / output | 1.1M |
| 10 | GPT-5.6 Sol· OpenAI | 1058±9 | 4 - 21 | $8.00 / $30.00 input / output | 1.1M |
| 11 | Gemini 3.5 Flash Lite· Google | 1057±9 | 4 - 22 | $0.30 / $2.50 input / output | 1M |
| 12 | GPT-6 Luna· OpenAI | 1051±38 | 1 - 45 | $0.20 / $0.75 input / output | 1.1M |
| 13 | Claude Opus 5.5· Anthropic | 1049±19 | 4 - 37 | $4.00 / $20.00 input / output | 1M |
| 14 | Claude Fable 5· Anthropic | 1049±9 | 7 - 29 | $10.00 / $50.00 input / output | 1M |
| 15 | Gemini 3 Flash· Google | 1046±10 | 7 - 32 | $0.50 / $3.00 input / output | 1M |
| 16 | GPT-5.4· OpenAI | 1044±10 | 8 - 34 | $5.00 / $22.50 input / output | 1M |
| 17 | GPT-5.6 Luna· OpenAI | 1044±10 | 9 - 36 | $0.40 / $1.80 input / output | 1.1M |
| 18 | GPT-5.6 Terra· OpenAI | 1043±10 | 9 - 36 | $4.00 / $18.00 input / output | 1.1M |
| 19 | Claude Opus 5· Anthropic | 1042±6 | 11 - 33 | $5.00 / $25.00 input / output | 1M |
| 20 | Qwen3.7 Max· Alibaba | 1041±9 | 9 - 37 | $2.50 / $7.50 input / output | 991K |
| 21 | Claude Sonnet 5· Anthropic | 1038±10 | 11 - 41 | $2.00 / $10.00 input / output | 1M |
| 22 | Qwen3.8 Max· Alibaba | 1038±10 | 11 - 41 | $2.00 / $6.00 input / output | 991K |
| 23 | Claude Opus 4.6· Anthropic | 1035±10 | 12 - 41 | $5.00 / $25.00 input / output | 1M |
| 24 | Claude Sonnet 4.6· Anthropic | 1034±10 | 12 - 41 | $3.00 / $15.00 input / output | 1M |
| 25 | Claude Opus 4.8· Anthropic | 1031±10 | 12 - 41 | $5.00 / $25.00 input / output | 1M |
| 26 | Gemini 3.1 Flash Lite· Google | 1029±10 | 13 - 41 | $0.25 / $1.50 input / output | 1M |
| 27 | Qwen3.8 Flash· Alibaba | 1028±21 | 10 - 50 | $0.15 / $0.47 input / output | 991K |
| 28 | Claude Opus 4.7· Anthropic | 1027±9 | 14 - 41 | $5.00 / $25.00 input / output | 1M |
| 29 | Mimo V2.6 Flash· Xiaomi | 1027±32 | 7 - 55 | $0.14 / $0.28 input / output | 1M |
| 30 | Grok-4.20 Multi-Agent· xAI | 1026±10 | 15 - 42 | $2.50 / $5.00 input / output | 2M |
| 31 | Qwen3.7 Plus· Alibaba | 1025±10 | 16 - 42 | $1.20 / $4.80 input / output | 1M |
| 32 | Grok-4.6· xAI | 1024±17 | 13 - 51 | $4.00 / $12.00 input / output | 500K |
| 33 | Qwen3.6 Plus· Alibaba | 1021±9 | 20 - 45 | $2.00 / $6.00 input / output | 1M |
| 34 | Qwen3.6 Max· Alibaba | 1019±10 | 20 - 46 | $2.00 / $12.00 input / output | 240K |
| 35 | Muse Spark 1.2· Meta | 1014±25 | 13 - 57 | $1.25 / $4.25 input / output | 1M |
| 36 | Claude Haiku 5.5· Anthropic | 1014±22 | 16 - 56 | $0.50 / $2.50 input / output | 1M |
| 37 | GLM 5.3· Z.ai | 1012±22 | 18 - 57 | $1.40 / $4.40 input / output | 1M |
| 38 | Claude Sonnet 5.5· Anthropic | 1010±47 | 7 - 61 | $2.00 / $10.00 input / output | 1M |
| 39 | Mimo V2.6 Pro· Xiaomi | 1009±38 | 11 - 60 | $0.44 / $0.87 input / output | 1M |
| 40 | Qwen3.8 27B· Alibaba | 1008±21 | 20 - 57 | $0.20 / $2.50 input / output | 262K |
| 41 | Muse Spark 1.1· Meta | 1007±12 | 27 - 55 | $1.25 / $4.25 input / output | 1M |
| 42 | Muse Spark 1.3· Meta | 1006±25 | 20 - 57 | $1.25 / $4.25 input / output | 1M |
| 43 | Kimi K3· Moonshot AI | 1005±11 | 29 - 55 | $3.00 / $15.00 input / output | 1M |
| 44 | Kimi K2.6· Moonshot AI | 1001±10 | 31 - 56 | $0.95 / $4.00 input / output | 262K |
| 45 | GLM 5.1· Z.ai | 1000±10 | 32 - 56 | $1.40 / $4.40 input / output | 203K |
| 46 | Grok-4.20 Reasoning· xAI | 1000±11 | 32 - 57 | $2.50 / $5.00 input / output | 2M |
| 47 | GLM 5.2· Z.ai | 999±10 | 32 - 57 | $1.40 / $4.40 input / output | 1M |
| 48 | DeepSeek V4 Pro 0813· DeepSeek | 997±19 | 29 - 58 | $1.32 / $3.96 input / output | 1M |
| 49 | GLM 5· Z.ai | 997±10 | 34 - 57 | $1.00 / $3.20 input / output | 203K |
| 50 | Mimo V2.5 Pro· Xiaomi | 993±10 | 34 - 57 | $0.44 / $0.87 input / output | 1M |
| 51 | DeepSeek V4 Flash 0731· DeepSeek | 989±13 | 34 - 59 | $0.14 / $0.28 input / output | 1M |
| 52 | GLM 5.3 Flash· Z.ai | 988±22 | 32 - 61 | $0.15 / $0.50 input / output | 1M |
| 53 | GLM 5 Turbo· Z.ai | 987±11 | 34 - 59 | $1.20 / $4.00 input / output | 203K |
| 54 | Mimo V2.5· Xiaomi | 983±11 | 37 - 60 | $0.14 / $0.28 input / output | 262K |
| 55 | DeepSeek V4 Flash· DeepSeek | 980±11 | 40 - 61 | $0.14 / $0.28 input / output | 1M |
| 56 | Mistral Medium 3.5· Mistral | 972±44 | 29 - 69 | $1.50 / $7.50 input / output | 262K |
| 57 | Kimi K2.5· Moonshot AI | 968±11 | 48 - 63 | $0.60 / $3.00 input / output | 262K |
| 58 | Grok-4.5· xAI | 966±13 | 49 - 63 | $4.00 / $12.00 input / output | 500K |
| 59 | Inkling· Thinkingmachines | 964±12 | 51 - 63 | $1.00 / $4.05 input / output | 256K |
| 60 | Grok-4.7· xAI | 962±46 | 33 - 70 | $2.00 / $6.00 input / output | 500K |
| 61 | GLM 5v Turbo· Z.ai | 960±11 | 53 - 63 | $1.20 / $4.00 input / output | 200K |
| 62 | Minimax M3· MiniMax | 950±12 | 56 - 65 | $0.30 / $1.20 input / output | 512K |
| 63 | Minimax M2.5 HighSpeed· MiniMax | 933±12 | 60 - 70 | $0.60 / $2.40 input / output | 205K |
| 64 | DeepSeek V3.2· DeepSeek | 932±12 | 60 - 70 | $0.62 / $1.85 input / output | 128K |
| 65 | DeepSeek V4.1 Flash· DeepSeek | 930±29 | 56 - 71 | $0.30 / $1.20 input / output | 1M |
| 66 | Minimax M2.7· MiniMax | 926±12 | 61 - 70 | $0.30 / $1.20 input / output | 205K |
| 67 | Mistral Large 3· Mistral | 925±12 | 61 - 70 | $0.50 / $1.50 input / output | 256K |
| 68 | DeepSeek V3.2 Thinking· DeepSeek | 921±12 | 61 - 70 | $0.62 / $1.85 input / output | 128K |
| 69 | Minimax M2.5· MiniMax | 919±12 | 61 - 70 | $0.30 / $1.20 input / output | 205K |
| 70 | Minimax M2.7 HighSpeed· MiniMax | 917±12 | 62 - 71 | $0.60 / $2.40 input / output | 205K |
| 71 | Grok-4.3· xAI | 893±14 | 69 - 72 | $2.50 / $5.00 input / output | 1M |
| 72 | Nemotron 3 Super· NVIDIA | 870±13 | 71 - 73 | $0.15 / $0.65 input / output | 256K |
| 73 | Ministral 14b· Mistral | 856±14 | 72 - 73 | $0.20 / $0.20 input / output | 256K |
| 74 | Grok-4.20 Non-Reasoning· xAI | 810±16 | 74 | $2.50 / $5.00 input / output | 2M |