Capability ranking
Best AI for Math
Best Math Models
Updated 2026-08-11
Ranked models208
Vendors25
Open-weight models49
Official prices93
What this ranking says
A quick reading of the current math ranking before the full data table.
GPT-5.6-Sol leads this page at 100.0, 0.0 points ahead of MiniMax M3.
MiniMax M3 is the highest-ranked open-weight option at #2. Nova Micro has the lowest official input price among ranked models.
Math Ranking Leaders
The leading math scores, with the axis focused on the competitive range rather than zero.
Math Capability Profile
Compare math strength with each model's broader capability score.
ModelReasoningKnowledgeMathCodingInstruction
Math scoreOverall score
80 models with comparable dataOpen a model by selecting its logo
Math Model Ranking
Models are ordered by their math score.
| 1 | OP | GPT-5.6 Sol | 100.0 | 95.1 | 87.8 | 95.8 | 92.9 | 90.6 | 100.0% | 1.1M | Closed weights | $5 | $30 |
| 2 | MI | MiniMax M3 | 100.0 | 71.3 | 69.2 | 74.0 | 53.8 | 62.6 | 100.0% | 512K | Open weights | $0.30 | $1.2 |
| 3 | AN | Claude Mythos Preview | 99.7 | 93.6 | 99.8 | 100.0 | 92.7 | 93.3 | 100.0% | N/A | Closed weights | N/A | N/A |
| 4 | AN | Claude Opus 5 | 99.5 | 97.4 | 100.0 | 75.0 | 90.4 | N/A | 80.0% | 1M | Closed weights | $5 | $25 |
| 5 | AN | Claude Fable 5 | 97.8 | 95.8 | N/A | 100.0 | 94.6 | 89.2 | 80.0% | 1M | Closed weights | $10 | $50 |
| 6 | GO | Gemini 2.0 Flash | 97.2 | 28.1 | 33.7 | 54.8 | 27.8 | 61.0 | 100.0% | 1M | Closed weights | N/A | N/A |
| 7 | ME | Muse Spark 1.1 | 96.7 | 88.3 | 87.5 | N/A | 75.7 | 89.4 | 80.0% | 1M | Closed weights | $1.25 | $4.25 |
| 8 | MA | Kimi K3 | 95.7 | 93.0 | 97.0 | 50.0 | 94.1 | 97.9 | 100.0% | 1M | Open weights | $3 | $15 |
| 9 | ME | Muse Spark | 95.7 | 73.2 | 78.9 | 100.0 | 49.4 | 89.8 | 100.0% | N/A | Closed weights | N/A | N/A |
| 10 | AN | Claude Opus 4.8 | 94.6 | 87.9 | 99.0 | 37.5 | 85.0 | 84.7 | 100.0% | 1M | Closed weights | $5 | $25 |
| 11 | ZA | GLM-5.2 | 89.5 | 79.3 | 92.7 | N/A | 56.5 | 100.0 | 80.0% | 1M | Open weights | $1.4 | $4.4 |
| 12 | AN | Claude Opus 4.7 | 88.6 | 84.2 | 98.7 | N/A | 69.3 | 89.5 | 80.0% | 1M | Closed weights | $5 | $25 |
| 13 | NV | Nemotron Nano 9B v2 | 87.1 | 32.4 | 34.3 | N/A | 75.0 | 52.9 | 80.0% | 131.1K | Open weights | N/A | N/A |
| 14 | CO | Command A+ | 86.8 | 41.9 | 33.1 | 67.7 | 33.3 | 51.8 | 100.0% | 128K | Open weights | $2.5 | $10 |
| 15 | DE | DeepSeek-V4-Pro-Max | 86.0 | 73.8 | 90.1 | N/A | 63.2 | 85.4 | 80.0% | 1M | Closed weights | $0.435 | $0.87 |
| 16 | GO | Gemini 2.0 Flash-Lite | 86.0 | 21.0 | 26.6 | 43.5 | N/A | 39.7 | 80.0% | 1M | Closed weights | N/A | N/A |
| 17 | AN | Claude Opus 4.6 | 85.9 | 79.6 | 77.5 | 50.0 | 72.2 | 77.2 | 100.0% | 1M | Closed weights | $5 | $25 |
| 18 | OP | GPT-4.5 | 85.0 | 36.2 | 43.8 | 69.3 | 40.5 | 52.8 | 100.0% | 128K | Closed weights | N/A | N/A |
| 19 | XI | MiMo-V2.5-Pro | 82.7 | 46.7 | 70.0 | 88.2 | 50.1 | 72.7 | 100.0% | 1M | Open weights | $0.435 | $0.87 |
| 20 | GO | Gemini 3.1 Pro | 82.6 | 77.5 | 91.8 | N/A | 55.9 | 74.1 | 80.0% | 1M | Closed weights | $2 | $12 |
| 21 | AC | Qwen2.5 32B Instruct | 81.1 | 14.8 | 30.0 | N/A | 68.3 | 53.6 | 80.0% | N/A | Closed weights | $0.70 | $2.8 |
| 22 | AM Nova ProAmazon | Nova Pro | 80.7 | 20.9 | 65.3 | 66.1 | 52.7 | 75.0 | 100.0% | 300K | Closed weights | $0.80 | $3.2 |
| 23 | OP | GPT-5.5 Pro | 80.6 | 75.8 | N/A | 50.0 | 94.7 | N/A | 60.0% | 1.1M | Closed weights | $30 | $180 |
| 24 | AC | Qwen2.5 72B Instruct | 80.1 | 20.9 | 22.5 | N/A | 60.3 | 54.9 | 80.0% | 131.1K | Closed weights | $1.4 | $5.6 |
| 25 | OP | GPT-5.2 Pro | 79.9 | 71.6 | 78.5 | 66.7 | 63.2 | 98.2 | 100.0% | 400K | Closed weights | $21 | $168 |
| 26 | AC | Qwen3.5-27B | 78.5 | 57.2 | 59.4 | 68.6 | 44.5 | 69.3 | 100.0% | 262.1K | Open weights | $0.30 | $2.4 |
| 27 | AN | Claude Sonnet 4.6 | 78.3 | 72.6 | 70.1 | N/A | 62.2 | 72.3 | 80.0% | 1M | Closed weights | $3 | $15 |
| 28 | ME | Llama 3.1 70B Instruct | 77.6 | 13.2 | 51.8 | N/A | 61.2 | 41.9 | 80.0% | 128K | Closed weights | N/A | N/A |
| 29 | AC | Qwen3.7 Max | 77.4 | 78.8 | 95.7 | N/A | 83.1 | 89.6 | 80.0% | 1M | Closed weights | $2.5 | $7.5 |
| 30 | XA | Grok-4 Heavy | 76.8 | 68.2 | 88.2 | N/A | 86.1 | 98.2 | 80.0% | N/A | Closed weights | N/A | N/A |
| 31 | ME | Llama 3.1 405B Instruct | 76.5 | 24.5 | 61.1 | N/A | 87.7 | 68.0 | 80.0% | 128K | Closed weights | N/A | N/A |
| 32 | AN | Claude 3.5 Sonnet | 76.1 | 32.1 | 80.4 | 45.2 | 48.8 | 80.4 | 100.0% | 200K | Closed weights | N/A | N/A |
| 33 | OP | GPT-5.2 | 76.1 | 71.4 | 75.5 | 60.7 | 59.1 | 85.6 | 100.0% | 400K | Closed weights | $1.75 | $14 |
| 34 | NV | Llama 3.1 Nemotron Ultra 253B v1 | 75.8 | 30.5 | 57.1 | N/A | 66.7 | 48.1 | 80.0% | N/A | Closed weights | N/A | N/A |
| 35 | AC | Qwen3.6 Plus | 74.4 | 67.5 | 84.9 | 82.0 | 48.8 | 77.3 | 100.0% | 1M | Closed weights | $0.50 | $3 |
| 36 | DE | DeepSeek-V3.2-Speciale | 74.0 | 58.1 | N/A | N/A | 46.1 | 83.2 | 60.0% | 131.1K | Closed weights | $0.14 | $0.28 |
| 37 | MA | Kimi K2.5 | 73.6 | 66.0 | 69.0 | 46.0 | 48.1 | 60.3 | 100.0% | 262.1K | Open weights | $0.60 | $3 |
| 38 | AC | Qwen3.5-122B-A10B | 73.4 | 59.2 | 61.7 | 87.8 | 51.6 | 68.5 | 100.0% | 262.1K | Open weights | $0.40 | $3.2 |
| 39 | OP | GPT-5.5 | 73.1 | 84.0 | 98.8 | 100.0 | 73.8 | 72.2 | 100.0% | 1.1M | Closed weights | $5 | $30 |
| 40 | ST | Step3-VL-10B | 72.2 | 43.6 | N/A | 75.8 | N/A | 77.2 | 60.0% | N/A | Closed weights | N/A | N/A |
| 41 | OP | GPT-5.4 | 72.2 | 77.1 | 90.3 | 83.3 | 57.8 | 71.2 | 100.0% | 1.1M | Closed weights | $2.5 | $15 |
| 42 | OP | GPT-5.6 Terra | 71.9 | 87.6 | 71.4 | 50.0 | 79.1 | 66.0 | 100.0% | 1.1M | Closed weights | $2 | $12 |
| 43 | AN | Claude Sonnet 5 | 71.7 | 84.0 | N/A | 62.5 | 78.8 | 87.2 | 80.0% | 1M | Closed weights | $2 | $10 |
| 44 | AC | Qwen3.8 Max | 70.7 | 89.3 | 97.8 | N/A | 85.8 | 97.9 | 80.0% | 1M | Closed weights | $2 | $6 |
| 45 | MA | Kimi-k1.5 | 70.3 | 30.1 | N/A | 49.2 | N/A | 62.0 | 60.0% | N/A | Closed weights | N/A | N/A |
| 46 | OP | GPT-4 Turbo | 70.3 | 14.6 | 21.5 | N/A | 55.4 | 69.2 | 80.0% | 128K | Closed weights | $10 | $30 |
| 47 | MA | Mistral Large 3 (675B Instruct 2512) | 70.0 | 15.8 | 16.9 | N/A | 24.3 | 36.4 | 80.0% | 262.1K | Open weights | $0.50 | $1.5 |
| 48 | GO | Gemini 3 Flash | 69.6 | 64.8 | 71.1 | 96.0 | 46.1 | 57.9 | 100.0% | 1M | Closed weights | $0.50 | $3 |
| 49 | DE | DeepSeek-V2.5 | 69.4 | 15.2 | N/A | N/A | 38.2 | 60.9 | 60.0% | 8.2K | Closed weights | N/A | N/A |
| 50 | ME | LongCat-Flash-Lite | 69.3 | 33.0 | 40.3 | N/A | 29.6 | 37.5 | 80.0% | 256K | Closed weights | N/A | N/A |
| 51 | GO | Gemma 3 27B | 69.1 | 15.8 | 20.4 | 20.0 | 39.2 | 50.1 | 100.0% | 131.1K | Closed weights | N/A | N/A |
| 52 | MA | Kimi K2.6 | 69.0 | 75.5 | 78.5 | N/A | 70.5 | 71.5 | 80.0% | 262.1K | Open weights | $0.95 | $4 |
| 53 | ME | Llama 3.3 70B Instruct | 68.6 | 22.0 | 24.9 | N/A | 66.9 | 68.7 | 80.0% | 128K | Open weights | N/A | N/A |
| 54 | ZA | GLM-5.1 | 68.0 | 69.6 | 80.7 | N/A | 47.1 | 54.8 | 80.0% | 200K | Open weights | $1.4 | $4.4 |
| 55 | BY | Seed 2.1 Turbo | 66.6 | 73.1 | 52.6 | 41.7 | 53.5 | 87.5 | 100.0% | N/A | Closed weights | N/A | N/A |
| 56 | OP | o1 | 66.3 | 36.3 | 59.4 | 72.6 | 46.7 | 56.3 | 100.0% | 200K | Closed weights | $15 | $60 |
| 57 | GO | Gemini 3.5 Flash | 65.2 | 79.2 | 90.0 | N/A | 64.5 | 60.5 | 80.0% | 1M | Closed weights | $1.5 | $9 |
| 58 | NV | Llama-3.3 Nemotron Super 49B v1 | 64.5 | 22.4 | 39.5 | N/A | 96.9 | 48.0 | 80.0% | 131.1K | Open weights | N/A | N/A |
| 59 | MA | Mistral Large 2 | 63.8 | 7.1 | N/A | N/A | 86.9 | 54.5 | 60.0% | 128K | Closed weights | N/A | N/A |
| 60 | MA | Pixtral Large | 63.2 | 21.2 | 82.1 | 38.7 | N/A | 65.6 | 80.0% | 128K | Closed weights | N/A | N/A |
| 61 | XA | Grok-2 | 62.4 | 21.0 | 29.2 | 41.9 | 66.9 | 45.3 | 100.0% | 128K | Closed weights | N/A | N/A |
| 62 | AC | Qwen3.7-Plus | 62.4 | 73.8 | 72.1 | 92.0 | 59.0 | 90.1 | 100.0% | 1M | Closed weights | $0.50 | $3 |
| 63 | GO | Gemma 4 31B | 62.1 | 60.1 | 69.5 | 63.6 | 71.7 | 76.0 | 100.0% | 262.1K | Open weights | N/A | N/A |
| 64 | AC | Qwen2.5 14B Instruct | 62.1 | 8.2 | 21.4 | N/A | 51.2 | 32.7 | 80.0% | N/A | Closed weights | $0.35 | $1.4 |
| 65 | BA | ERNIE 5.0 | 62.0 | 58.7 | 77.3 | N/A | N/A | 78.8 | 60.0% | N/A | Closed weights | N/A | N/A |
| 66 | AC | Qwen2 72B Instruct | 61.0 | 9.8 | 22.8 | N/A | 46.6 | 51.9 | 80.0% | N/A | Closed weights | N/A | N/A |
| 67 | NV | Nemotron 3 Super (120B A12B) | 60.8 | 44.8 | 70.0 | N/A | 23.5 | 64.5 | 80.0% | 262.1K | Open weights | $0.20 | $0.80 |
| 68 | AC | Qwen3 VL 32B Thinking | 60.6 | 40.5 | 65.5 | 64.8 | 60.2 | 58.1 | 100.0% | N/A | Closed weights | N/A | N/A |
| 69 | AN | Claude 3 Opus | 60.5 | 16.7 | 59.2 | N/A | 46.1 | 72.7 | 80.0% | 200K | Closed weights | N/A | N/A |
| 70 | MA | Kimi K2-Thinking-0905 | 60.2 | 61.0 | 75.5 | N/A | 55.0 | 67.0 | 80.0% | 262.1K | Closed weights | $0.60 | $2.5 |
| 71 | AMNova LiteAmazon | Nova Lite | 59.9 | 13.2 | 51.1 | 35.5 | 34.2 | 42.6 | 100.0% | 300K | Closed weights | $0.06 | $0.24 |
| 72 | AC | Qwen3.5-35B-A3B | 59.2 | 52.0 | 39.3 | 52.4 | 42.7 | 58.2 | 100.0% | 262.1K | Open weights | $0.25 | $2 |
| 73 | BY | Seed 2.1 Pro | 58.8 | 78.7 | 80.1 | 91.7 | 64.3 | 93.1 | 100.0% | N/A | Closed weights | N/A | N/A |
| 74 | ME | Llama 4 Maverick | 58.6 | 24.9 | 53.8 | 61.3 | 43.2 | 79.5 | 100.0% | 1M | Closed weights | N/A | N/A |
| 75 | AC | Qwen2.5 7B Instruct | 58.4 | 6.5 | 9.4 | N/A | 37.6 | 24.6 | 80.0% | 131.1K | Closed weights | $0.175 | $0.70 |
| 76 | GO | Gemma 3 12B | 57.3 | 11.2 | 14.1 | 10.0 | 32.4 | 51.9 | 100.0% | 131.1K | Closed weights | N/A | N/A |
| 77 | OP | o4-mini | 57.2 | 45.0 | 66.7 | 82.3 | 52.2 | 44.5 | 100.0% | 200K | Closed weights | $1.1 | $4.4 |
| 78 | ZA | GLM-4.5 | 56.8 | 49.3 | 61.4 | N/A | 58.4 | 84.5 | 80.0% | 131.1K | Open weights | $0.60 | $2.2 |
| 79 | AN | Claude 3.5 Haiku | 56.8 | 9.3 | 13.7 | N/A | 19.9 | 38.0 | 80.0% | 200K | Closed weights | N/A | N/A |
| 80 | AN | Claude 3.7 Sonnet | 56.5 | 38.7 | 76.8 | 66.1 | 60.5 | 51.3 | 100.0% | 200K | Closed weights | N/A | N/A |
| 81 | OP | o3-mini | 56.3 | 36.3 | 58.1 | N/A | 36.2 | 63.9 | 80.0% | 200K | Closed weights | $1.1 | $4.4 |
| 82 | XA | Grok-2 mini | 56.1 | 16.3 | 26.0 | 37.1 | 50.8 | 39.8 | 100.0% | N/A | Closed weights | N/A | N/A |
| 83 | MI | Phi 4 | 56.1 | 9.4 | 29.6 | N/A | 38.5 | 18.1 | 80.0% | 16K | Closed weights | $0.125 | $0.50 |
| 84 | ST | Step-3.5-Flash | 55.6 | 60.4 | N/A | N/A | 57.8 | 88.5 | 60.0% | 256K | Open weights | $0.10 | $0.30 |
| 85 | AC | Qwen3.6-27B | 54.8 | 60.5 | 79.4 | 83.8 | 45.5 | 72.7 | 100.0% | 262.1K | Open weights | $0.60 | $3.6 |
| 86 | ME | LongCat-Flash-Chat | 54.8 | 36.9 | 50.2 | N/A | 42.8 | 55.5 | 80.0% | 128K | Closed weights | N/A | N/A |
| 87 | AC | Qwen3.5-397B-A17B | 53.7 | 66.3 | 90.8 | N/A | 69.8 | 80.0 | 80.0% | 262.1K | Open weights | $0.60 | $3.6 |
| 88 | ZA | GLM-4.7 | 53.7 | 59.7 | 79.4 | N/A | 47.9 | 81.4 | 80.0% | 204.8K | Open weights | $0.60 | $2.2 |
| 89 | NV | Nemotron 3 Ultra (550B A55B) | 53.6 | 62.9 | 82.6 | N/A | 46.0 | 89.9 | 80.0% | 1M | Open weights | $0.50 | $2.5 |
| 90 | NV | Llama 3.1 Nemotron 70B Instruct | 53.2 | 3.1 | 33.3 | N/A | N/A | 48.5 | 60.0% | 128K | Open weights | N/A | N/A |
| 91 | GO | Gemini 1.5 Pro | 52.0 | 20.5 | 31.3 | 40.3 | 41.5 | 72.8 | 100.0% | 2.1M | Closed weights | N/A | N/A |
| 92 | ZA | GLM-4.5-Air | 51.1 | 41.5 | 54.3 | N/A | 48.2 | 74.9 | 80.0% | 131.1K | Open weights | $0.20 | $1.1 |
| 93 | OP | GPT-5.4 mini | 51.1 | 60.5 | 72.5 | N/A | 37.7 | 54.4 | 80.0% | 400K | Closed weights | $0.75 | $4.5 |
| 94 | XA | Grok 4 Fast | 50.5 | 54.0 | 79.4 | N/A | 53.2 | 82.1 | 80.0% | 2M | Closed weights | N/A | N/A |
| 95 | OP | GPT-5.1 | 50.0 | 62.1 | 86.5 | 96.8 | 70.9 | 77.9 | 100.0% | 400K | Closed weights | $1.25 | $10 |
| 96 | OP | GPT-5.1 Instant | 50.0 | 61.3 | 86.5 | 96.8 | 70.9 | 77.9 | 100.0% | 400K | Closed weights | N/A | N/A |
| 97 | SA | Sarvam-105B | 49.9 | 44.6 | 60.1 | N/A | 28.8 | 65.1 | 80.0% | 131.1K | Open weights | N/A | N/A |
| 98 | AC | Qwen3 VL 32B Instruct | 48.9 | 35.2 | 58.8 | 70.6 | 46.4 | 43.9 | 100.0% | N/A | Closed weights | N/A | N/A |
| 99 | AC | Qwen3 VL 235B A22B Thinking | 48.7 | 45.8 | 53.9 | 45.3 | 61.6 | 62.2 | 100.0% | 262.1K | Closed weights | N/A | N/A |
| 100 | MA | Mistral Small 3 24B Instruct | 48.6 | 7.5 | 18.0 | N/A | 43.9 | 29.1 | 80.0% | 32K | Closed weights | N/A | N/A |
| 101 | XA | Grok-4 | 47.8 | 56.5 | 51.9 | N/A | 83.3 | 65.5 | 80.0% | 256K | Closed weights | N/A | N/A |
| 102 | AC | Qwen3.6-35B-A3B | 47.4 | 55.3 | 57.0 | 74.0 | 26.5 | 72.1 | 100.0% | 262.1K | Open weights | $0.248 | $1.49 |
| 103 | OP | o3 | 47.3 | 52.6 | 50.1 | 63.8 | 51.0 | 62.1 | 100.0% | 200K | Closed weights | $2 | $8 |
| 104 | TE | Hy3 | 47.2 | 71.0 | 45.7 | 0.0 | 63.0 | N/A | 80.0% | 256K | Open weights | N/A | N/A |
| 105 | GO | Gemma 4 26B-A4B | 47.2 | 52.6 | 57.3 | 54.5 | 67.5 | 63.8 | 100.0% | 262.1K | Open weights | N/A | N/A |
| 106 | AC | Qwen3-235B-A22B-Thinking-2507 | 46.6 | 48.5 | 71.4 | N/A | 40.4 | 80.0 | 80.0% | 262.1K | Closed weights | N/A | N/A |
| 107 | AC | Qwen2.5-Coder 32B Instruct | 46.2 | 4.4 | 39.4 | N/A | 68.0 | 33.5 | 80.0% | 128K | Closed weights | $0.287 | $0.861 |
| 108 | MA | Mistral Small 3.2 24B Instruct | 45.8 | 15.9 | 57.7 | 35.5 | N/A | 36.4 | 80.0% | N/A | Closed weights | N/A | N/A |
| 109 | OP | GPT-5.4 nano | 45.6 | 49.1 | 54.5 | N/A | 19.8 | 35.6 | 80.0% | 400K | Closed weights | $0.20 | $1.25 |
| 110 | NV | Llama 3.1 Nemotron Nano 8B V1 | 45.2 | 10.4 | 28.8 | N/A | 84.4 | 14.7 | 80.0% | N/A | Closed weights | N/A | N/A |
| 111 | AC | Qwen3 VL 235B A22B Instruct | 44.6 | 43.2 | 65.5 | 45.3 | 59.3 | 62.5 | 100.0% | 262.1K | Closed weights | N/A | N/A |
| 112 | ME | Llama 4 Scout | 44.3 | 14.1 | 56.3 | 46.8 | 23.8 | 60.0 | 100.0% | 10M | Closed weights | N/A | N/A |
| 113 | OP | GPT-5.6 Luna | 43.8 | 76.4 | 55.2 | 25.0 | 68.6 | 40.8 | 100.0% | 1.1M | Closed weights | $0.20 | $1.2 |
| 114 | AC | Qwen3 VL 30B A3B Thinking | 43.3 | 31.8 | 45.6 | 52.3 | 38.8 | 39.6 | 100.0% | 262.1K | Closed weights | N/A | N/A |
| 115 | OP | GPT-4o mini | 42.9 | 8.1 | 12.0 | 29.0 | 28.5 | 55.3 | 100.0% | 128K | Closed weights | $0.15 | $0.60 |
| 116 | MA | Kimi K2-Instruct-0905 | 42.5 | 37.2 | 55.1 | N/A | 31.7 | 50.0 | 80.0% | N/A | Closed weights | $0.60 | $2.5 |
| 117 | ME | Muse Glimmer-30B | 40.8 | 58.5 | 60.1 | N/A | 45.3 | 62.9 | 80.0% | 131.1K | Open weights | N/A | N/A |
| 118 | MI | MiniMax M2.1 | 40.8 | 53.7 | 65.2 | N/A | 51.5 | 58.9 | 80.0% | 204.8K | Open weights | $0.30 | $1.2 |
| 119 | MA | Mistral Small 3.1 24B Instruct | 40.7 | 8.4 | 18.9 | 53.0 | 59.2 | 19.6 | 100.0% | N/A | Closed weights | N/A | N/A |
| 120 | GO | Gemini 1.5 Flash | 40.4 | 10.7 | 26.0 | 33.9 | 23.1 | 46.6 | 100.0% | 1M | Closed weights | N/A | N/A |
| 121 | GO | Gemini 2.5 Pro Preview 06-05 | 39.1 | 49.9 | 81.1 | 70.7 | 66.1 | 72.6 | 100.0% | 1M | Closed weights | $1.25 | $10 |
| 122 | MI | MiniMax M1 80K | 39.0 | 37.0 | 45.1 | N/A | 56.9 | 46.3 | 80.0% | 1M | Closed weights | N/A | N/A |
| 123 | GO | Gemma 3 4B | 38.9 | 0.0 | 1.7 | 0.0 | 10.8 | 26.2 | 100.0% | 131.1K | Closed weights | N/A | N/A |
| 124 | AN | Claude 3 Sonnet | 38.7 | 3.5 | 48.8 | N/A | 20.0 | 53.6 | 80.0% | 200K | Closed weights | N/A | N/A |
| 125 | AC | Qwen2.5 VL 32B Instruct | 38.6 | 17.9 | 32.0 | 49.2 | 49.4 | 15.2 | 100.0% | N/A | Closed weights | N/A | N/A |
| 126 | XI | MiMo-V2-Flash | 38.4 | 51.3 | 73.0 | N/A | 44.9 | 60.0 | 80.0% | 262.1K | Open weights | $0.14 | $0.28 |
| 127 | AMNova MicroAmazon | Nova Micro | 38.3 | 4.3 | 40.6 | 0.0 | 17.3 | 13.0 | 100.0% | 128K | Closed weights | $0.035 | $0.14 |
| 128 | AC | Qwen3 VL 8B Thinking | 38.2 | 28.5 | 39.3 | 30.8 | 28.8 | 38.3 | 100.0% | 262.1K | Closed weights | N/A | N/A |
| 129 | AC | Qwen3-Next-80B-A3B-Thinking | 37.5 | 40.9 | 58.2 | N/A | 41.2 | 64.5 | 80.0% | 131.1K | Open weights | $0.50 | $6 |
| 130 | GO | Gemma 3n E4B Instructed | 37.5 | 0.0 | 8.0 | N/A | 16.9 | 38.2 | 80.0% | 32K | Closed weights | N/A | N/A |
| 131 | MI | Phi-4-multimodal-instruct | 36.8 | 7.4 | 25.6 | 17.7 | N/A | 25.0 | 80.0% | 128K | Closed weights | N/A | N/A |
| 132 | GO | Gemini 3 Pro | 36.4 | 67.3 | 68.8 | 100.0 | 60.5 | 64.3 | 100.0% | 1M | Closed weights | N/A | N/A |
| 133 | OP | GPT-4 | 36.3 | 4.6 | 8.6 | N/A | 10.8 | 74.9 | 80.0% | 8.2K | Closed weights | $30 | $60 |
| 134 | MI | Phi 4 Mini | 35.6 | 0.0 | 45.0 | 62.5 | N/A | 25.1 | 80.0% | 128K | Open weights | $0.075 | $0.30 |
| 135 | AC | Qwen2.5-Omni-7B | 35.3 | 10.2 | 21.8 | 27.7 | 26.1 | 25.1 | 100.0% | N/A | Closed weights | $0.10 | $0.40 |
| 136 | MI | Phi-3.5-MoE-instruct | 34.3 | 4.9 | 60.9 | 100.0 | 39.7 | 57.6 | 100.0% | N/A | Closed weights | N/A | N/A |
| 137 | NR | Hermes 3 70B | 34.0 | 8.0 | 29.7 | 35.7 | N/A | 68.8 | 80.0% | 131.1K | Closed weights | N/A | N/A |
| 138 | GO | DiffusionGemma 26B-A4B | 34.0 | 34.2 | 33.4 | 45.5 | 18.6 | 40.5 | 100.0% | N/A | Closed weights | N/A | N/A |
| 139 | OP | GPT-5 mini | 33.4 | 44.9 | 68.5 | N/A | N/A | 63.3 | 60.0% | 400K | Closed weights | $0.25 | $2 |
| 140 | GO | Gemma 4 12B | 32.8 | 38.2 | 51.8 | 36.4 | 29.3 | 41.9 | 100.0% | N/A | Closed weights | N/A | N/A |
| 141 | ME | Llama 3.2 90B Instruct | 32.5 | 10.8 | 35.0 | 30.6 | N/A | 41.3 | 80.0% | 128K | Closed weights | N/A | N/A |
| 142 | DE | DeepSeek-V3 0324 | 32.3 | 26.0 | 42.1 | N/A | 36.1 | 41.9 | 80.0% | 163.8K | Closed weights | N/A | N/A |
| 143 | AL | Jamba 1.5 Large | 31.9 | 2.7 | 46.1 | N/A | N/A | 27.4 | 60.0% | 256K | Closed weights | N/A | N/A |
| 144 | GO | Gemma 3n E4B Instructed LiteRT Preview | 31.7 | 0.0 | 21.1 | 41.2 | 16.9 | 29.0 | 100.0% | N/A | Closed weights | N/A | N/A |
| 145 | SA | Sarvam-30B | 30.5 | 34.3 | 39.1 | N/A | 49.6 | 71.8 | 80.0% | 128K | Open weights | N/A | N/A |
| 146 | MI | MAI-Thinking-1 | 30.0 | 55.1 | 74.5 | 13.6 | 54.4 | 75.1 | 100.0% | N/A | Closed weights | N/A | N/A |
| 147 | ZA | GLM-4.6 | 29.9 | 54.2 | 65.2 | N/A | 50.4 | 76.1 | 80.0% | 204.8K | Open weights | $0.60 | $2.2 |
| 148 | MI | MiniMax M1 40K | 29.6 | 35.3 | 43.4 | N/A | 50.9 | 48.3 | 80.0% | N/A | Closed weights | N/A | N/A |
| 149 | AN | Claude 3 Haiku | 29.2 | 0.0 | 37.0 | N/A | 27.7 | 44.5 | 80.0% | 200K | Closed weights | N/A | N/A |
| 150 | AC | Qwen3 VL 30B A3B Instruct | 28.7 | 28.8 | 31.7 | 44.0 | 26.8 | 36.7 | 100.0% | 262.1K | Closed weights | N/A | N/A |
| 151 | ME | LongCat-Flash-Thinking-2601 | 28.6 | 60.8 | 63.5 | N/A | 76.2 | 93.4 | 80.0% | 128K | Closed weights | N/A | N/A |
| 152 | OP | GPT-4.1 mini | 28.4 | 24.8 | 35.6 | 58.1 | 13.1 | 30.9 | 100.0% | 1M | Closed weights | $0.40 | $1.6 |
| 153 | DE | DeepSeek-R1-0528 | 28.3 | 44.2 | 65.2 | N/A | 22.5 | 77.2 | 80.0% | 131.1K | Closed weights | N/A | N/A |
| 154 | AC | Qwen3.5-9B | 28.1 | 41.4 | 58.9 | N/A | 33.5 | 46.6 | 80.0% | 262.1K | Open weights | N/A | N/A |
| 155 | AC | Qwen3 VL 4B Thinking | 27.5 | 23.0 | 27.2 | 28.3 | 25.5 | 25.8 | 100.0% | 262.1K | Closed weights | N/A | N/A |
| 156 | GO | Gemini 3.1 Flash-Lite | 27.2 | 51.9 | 71.8 | 80.0 | 10.3 | 50.1 | 100.0% | 1M | Closed weights | $0.25 | $1.5 |
| 157 | XA | Grok-1.5 | 26.9 | 0.0 | 9.0 | 13.7 | 21.5 | 13.3 | 100.0% | N/A | Closed weights | N/A | N/A |
| 158 | OP | GPT-4.1 | 26.7 | 31.9 | 38.6 | 64.5 | 37.7 | 39.0 | 100.0% | 1M | Closed weights | $2 | $8 |
| 159 | GO | Gemma 4 E4B | 25.8 | 24.4 | 22.2 | 18.2 | 18.2 | 31.7 | 100.0% | 131.1K | Open weights | N/A | N/A |
| 160 | DE | DeepSeek R1 Distill Qwen 7B | 25.8 | 15.7 | 23.2 | N/A | 29.9 | 61.5 | 80.0% | N/A | Closed weights | N/A | N/A |
| 161 | GO | Gemma 2 27B | 25.6 | 0.1 | 69.3 | 97.1 | 8.6 | 64.2 | 100.0% | N/A | Closed weights | N/A | N/A |
| 162 | MA | Mistral Medium 3.5 | 25.0 | 49.2 | N/A | N/A | 38.3 | 48.6 | 60.0% | 262.1K | Open weights | $1.5 | $7.5 |
| 163 | NV | Nemotron 3 Nano (30B A3B) | 25.0 | 35.2 | 54.3 | N/A | 9.7 | 65.2 | 80.0% | 262.1K | Open weights | N/A | N/A |
| 164 | GO | Gemini 1.5 Flash 8B | 24.5 | 0.0 | 11.2 | 16.1 | N/A | 13.1 | 80.0% | 1M | Closed weights | N/A | N/A |
| 165 | MI | Phi-3.5-mini-instruct | 24.4 | 0.0 | 50.7 | 62.5 | 18.7 | 31.5 | 100.0% | 128K | Closed weights | N/A | N/A |
| 166 | MA | Pixtral-12B | 23.8 | 0.0 | 28.3 | 11.3 | 16.9 | 20.0 | 100.0% | 128K | Open weights | $0.15 | $0.15 |
| 167 | OP | GPT-4.1 nano | 21.1 | 4.3 | 24.0 | 19.4 | 0.0 | 6.1 | 100.0% | 1M | Closed weights | $0.10 | $0.40 |
| 168 | AC | Qwen3 VL 8B Instruct | 21.0 | 24.8 | 39.2 | 16.0 | 29.3 | 27.1 | 100.0% | 262.1K | Closed weights | N/A | N/A |
| 169 | AC | QwQ-32B | 21.0 | 24.8 | 36.5 | N/A | 34.9 | 40.0 | 80.0% | N/A | Closed weights | $0.287 | $0.861 |
| 170 | AC | Qwen3-235B-A22B-Instruct-2507 | 20.8 | 42.4 | 61.1 | 0.0 | 40.2 | 65.8 | 100.0% | 262.1K | Open weights | $0.70 | $2.8 |
| 171 | AC | Qwen2.5 VL 72B Instruct | 20.8 | 22.7 | 42.5 | 44.6 | 12.6 | 37.2 | 100.0% | 131.1K | Open weights | N/A | N/A |
| 172 | ZA | GLM-4.7-Flash | 20.1 | 38.5 | 55.8 | N/A | 17.7 | 64.6 | 80.0% | 200K | Open weights | N/A | N/A |
| 173 | MI | MAI-Code-1-Flash | 20.0 | 50.4 | 76.2 | N/A | 31.1 | 55.7 | 80.0% | 256K | Closed weights | N/A | N/A |
| 174 | AC | Qwen2.5-Coder 7B Instruct | 19.9 | 0.0 | 12.1 | N/A | 50.5 | 16.0 | 80.0% | N/A | Closed weights | $0.144 | $0.287 |
| 175 | OP | GPT-4o | 19.8 | 23.9 | 42.9 | 30.2 | 10.8 | 23.7 | 100.0% | 128K | Closed weights | $2.5 | $10 |
| 176 | GO | Gemma 4 E2B | 19.4 | 11.6 | 11.0 | 9.1 | 8.1 | 19.9 | 100.0% | 131.1K | Open weights | N/A | N/A |
| 177 | MA | Ministral 8B Instruct | 19.1 | 0.0 | 48.5 | 17.6 | 0.0 | 26.1 | 100.0% | 131.1K | Open weights | N/A | N/A |
| 178 | AC | Qwen3.5-4B | 18.8 | 32.4 | 45.4 | N/A | 20.6 | 35.5 | 80.0% | N/A | Closed weights | N/A | N/A |
| 179 | GO | Gemma 2 9B | 18.4 | 0.0 | 48.0 | 72.3 | 2.3 | 41.3 | 100.0% | N/A | Closed weights | N/A | N/A |
| 180 | AC | Qwen2-VL-72B-Instruct | 18.2 | 25.5 | 57.1 | 34.4 | N/A | 81.3 | 80.0% | N/A | Closed weights | N/A | N/A |
| 181 | MI | MiniMax M2 | 17.4 | 43.9 | 59.4 | N/A | 41.4 | 52.4 | 80.0% | 196.6K | Open weights | $0.30 | $1.2 |
| 182 | OP | GPT-5 nano | 16.8 | 31.1 | 47.2 | N/A | N/A | 47.8 | 60.0% | 400K | Closed weights | $0.05 | $0.40 |
| 183 | ME | Llama 3.2 11B Instruct | 16.8 | 0.0 | 27.3 | 8.1 | N/A | 26.1 | 80.0% | 128K | Closed weights | N/A | N/A |
| 184 | ME | Llama 3.2 3B Instruct | 16.7 | 0.0 | 29.1 | N/A | N/A | 10.7 | 60.0% | 128K | Closed weights | N/A | N/A |
| 185 | AC | Qwen3-Next-80B-A3B-Instruct | 16.7 | 36.6 | 53.4 | N/A | 28.4 | 54.1 | 80.0% | 131.1K | Open weights | $0.50 | $2 |
| 186 | DE | DeepSeek VL2 Tiny | 15.8 | 0.0 | 16.6 | 1.6 | N/A | 0.0 | 80.0% | N/A | Closed weights | N/A | N/A |
| 187 | AC | Qwen2 7B Instruct | 14.9 | 0.0 | 3.0 | N/A | 17.1 | 17.3 | 80.0% | N/A | Closed weights | N/A | N/A |
| 188 | AL | Jamba 1.5 Mini | 14.9 | 0.0 | 34.8 | N/A | N/A | 7.5 | 60.0% | 256.1K | Closed weights | N/A | N/A |
| 189 | DE | DeepSeek-V3.1 | 14.6 | 39.9 | 53.6 | N/A | 31.4 | 61.6 | 80.0% | 131.1K | Open weights | N/A | N/A |
| 190 | GO | Gemini 2.5 Flash | 14.1 | 38.1 | 70.6 | 79.0 | 46.1 | 52.5 | 100.0% | 1M | Closed weights | $0.30 | $2.5 |
| 191 | OP | GPT-3.5 Turbo | 13.9 | 0.0 | 5.2 | 0.0 | 12.3 | 13.6 | 100.0% | 16.4K | Closed weights | $0.50 | $1.5 |
| 192 | IB | Granite 3.3 8B Instruct | 13.4 | 3.8 | N/A | N/A | 80.8 | 49.7 | 60.0% | 128K | Closed weights | N/A | N/A |
| 193 | AC | Qwen3 VL 4B Instruct | 12.9 | 19.3 | 20.8 | 12.1 | 17.8 | 19.6 | 100.0% | 262.1K | Closed weights | N/A | N/A |
| 194 | OP | o1-mini | 12.9 | 16.8 | 32.6 | N/A | 90.8 | 59.1 | 80.0% | 128K | Closed weights | N/A | N/A |
| 195 | GO | Gemma 3n E2B Instructed | 12.5 | 0.0 | 1.8 | N/A | 7.0 | 12.4 | 80.0% | N/A | Closed weights | N/A | N/A |
| 196 | GO | Gemma 3n E2B Instructed LiteRT (Preview) | 12.3 | 0.0 | 5.2 | 15.1 | 7.0 | 12.5 | 100.0% | N/A | Closed weights | N/A | N/A |
| 197 | XA | Grok-1.5V | 11.8 | 1.3 | 28.0 | 13.7 | N/A | N/A | 60.0% | N/A | Closed weights | N/A | N/A |
| 198 | MA | Magistral Medium | 10.9 | 19.8 | 46.4 | N/A | 29.7 | 18.6 | 80.0% | N/A | Closed weights | N/A | N/A |
| 199 | CO | Command R+ | 10.6 | 0.0 | 42.4 | N/A | N/A | 58.2 | 60.0% | 128K | Closed weights | $0.15 | $0.60 |
| 200 | ME | Llama 3.1 8B Instruct | 10.3 | 0.0 | 29.3 | N/A | 49.2 | 10.2 | 80.0% | 131.1K | Closed weights | N/A | N/A |
| 201 | DE | DeepSeek R1 Distill Llama 8B | 9.7 | 12.4 | 22.5 | N/A | 31.9 | 46.1 | 80.0% | N/A | Closed weights | N/A | N/A |
| 202 | AC | Qwen2.5 VL 7B Instruct | 8.4 | 12.8 | 37.6 | 23.8 | 6.7 | 16.9 | 100.0% | N/A | Closed weights | $0.35 | $1.05 |
| 203 | IB | IBM Granite 4.0 Tiny Preview | 7.7 | 0.0 | N/A | N/A | 36.9 | 13.2 | 60.0% | N/A | Closed weights | N/A | N/A |
| 204 | GO | Gemma 3 1B | 6.9 | 0.0 | 0.4 | N/A | 1.0 | 2.5 | 80.0% | N/A | Closed weights | N/A | N/A |
| 205 | GO | Gemini 1.0 Pro | 4.1 | 0.0 | 3.4 | 4.8 | N/A | 40.2 | 80.0% | 32.8K | Closed weights | N/A | N/A |
| 206 | GO | Gemini 2.5 Flash-Lite | 3.3 | 19.1 | 35.2 | 59.7 | 9.8 | 36.1 | 100.0% | 1M | Closed weights | $0.10 | $0.40 |
| 207 | MI | Phi-3.5-vision-instruct | 2.6 | 0.0 | 18.7 | 3.2 | N/A | 37.5 | 80.0% | N/A | Closed weights | N/A | N/A |
| 208 | BA | ERNIE 4.5 | 0.0 | 0.0 | 25.9 | 0.0 | 0.0 | 0.1 | 100.0% | 128K | Closed weights | N/A | N/A |
Official PAYG prices appear here. Third-party offers remain on the pricing and model detail pages.
Capability at each price point
Official vendor API prices plotted against the math metric used on this page. Price remains a separate decision signal.
Math scoreOfficial API price blend (8:1 input/output), USD / 1M
Efficient frontier93 models with official PAYG prices
Math Ranking Composition
Vendor concentration and model access among the first 25 ranked products.
Top-model vendor mix
25 modelsCompare the leaders
Open focused comparisons between the current leader and the nearest practical alternatives.
About this leaderboard
Common questions about Best AI for Math.
What is the best model on Best AI for Math?
GPT-5.6-Sol is currently ranked first with a math score of 100.0.
How is this leaderboard ranked?
Each model appears once using its current scored version. Category pages use the capability named in the title, while the overall page uses the LLMBoard score.
How many models are included?
This page currently ranks 208 unique model products.
Where does pricing come from?
Main price columns use the model vendor's official standard PAYG API rate. Eligible third-party offers appear only in separately labeled columns, and unavailable official prices display as N/A.
Does Arena affect the score?
No. Arena results are displayed as an independent signal and are not included in the current LLMBoard capability score.