Best models first. Then what they cost.
97 of 108 rated, priced models are beaten on quality and undercut on price by something else on this page. 11 are not.
Best for
Measured on
Workload
Prompt size
106 ranked· 221 unrated · 51 distinct ranks
| # | Notes | ||||
|---|---|---|---|---|---|
| 1 Rank 1, tied with 1 others — on the value frontier | Claude Opus 5 Anthropic | 63.1 1504.2arena | $10.00 $5.00in$25.00out | 1M | frontier R T 3M |
| 1 Rank 1, tied with 3 others | Claude Fable 5 Anthropic | 62.1 1494.7arena | $20.00 $10.00in$50.00out | 1M | R T 3M |
| 2 Rank 2, tied with 4 others — on the value frontier | Grok 4.6 xAI | 60.9 1443.7arena | $3.00 $2.00in$6.00out ×2 past 200K | 500K | frontier R T 3M |
| 2 Rank 2, tied with 4 others | GPT-5.6 Sol OpenAI | 60.9 1454.2arena | $4.00 $2.00in$10.00out ×2 past 272K | 1.1M | R T 3M |
| 3 Rank 3, tied with 3 others | Kimi K3 Moonshot AI | 59.7 1476.3arena | $6.00 $3.00in$15.00out | 1.0M | open R T 3M |
| 3 Rank 3, tied with 4 others — on the value frontier | GLM 5.3 Z.ai | 59.5 1476.9arena | $2.15 $1.40in$4.40out | 1.0M | frontier R T |
| 6 Rank 6, tied with 4 others | Qwen3.8 Max Qwen | 58.1 | $3.00 $2.00in$6.00out | 1M | R T 3M |
| 7 Rank 7, tied with 5 others | Qwen3.8 2.4T A95B Qwen | 57.7 | $3.00 $2.00in$6.00out | 1.0M | open R T |
| 7 Rank 7, tied with 6 others | Claude Opus 4.8 Anthropic | 57.3 1460.9arena | $10.00 $5.00in$25.00out | 1M | R T 3M |
| 7 Rank 7, tied with 7 others — on the value frontier | Muse Spark 1.2 Meta | 56.8 | $2.00 $1.25in$4.25out | 1.0M | frontier R T 5M |
| 8 Rank 8, tied with 7 others | GPT-5.6 Terra OpenAI | 56.6 1446arena | $4.50 $2.00in$12.00out ×2 past 272K | 1.1M | R T 3M |
| 8 Rank 8, tied with 8 others | GPT-5.5 OpenAI | 56.3 1470.9arena | $11.25 $5.00in$30.00out ×2 past 272K | 1.1M | R T 3M |
| 9 Rank 9, tied with 7 others — on the value frontier | Gemini 3.7 Flash Google | 56.0 1490.2arena | $0.750 $0.375in$1.88out + reasoning billed | 1.0M | frontier R T 5M |
| 10 Rank 10, tied with 6 others | Grok 4.5 xAI | 55.8 1452.3arena | $3.00 $2.00in$6.00out ×2 past 200K | 500K | R T 3M |
| 11 Rank 11, tied with 5 others | Claude Sonnet 5 Anthropic | 55.3 1442.3arena | $4.00 $2.00in$10.00out | 1M | R T 3M |
| 12 Rank 12, tied with 4 others | Claude Opus 4.7 Anthropic | 55.0 1490.1arena | $10.00 $5.00in$25.00out | 1M | R T 3M |
| 17 Rank 17, tied with 7 others | DeepSeek V4 Pro 0813 DeepSeek | 53.2 | $1.68 $1.12in$3.37out | 1.0M | R T |
| 17 Rank 17, tied with 7 others | Muse Spark 1.1 Meta | 53.2 1478.3arena | $2.00 $1.25in$4.25out | 1.0M | R T 5M |
| 17 Rank 17, tied with 7 others | GPT-5.4 OpenAI | 53.1 1469.6arena | $5.63 $2.50in$15.00out ×2 past 272K | 1.1M | R T 3M |
| 17 Rank 17, tied with 8 others | GLM 5.2 Z.ai | 52.6 1465.4arena | $1.48 $0.966in$3.04out | 1.0M | open R T |
| 17 Rank 17, tied with 8 others — on the value frontier | GPT-5.6 Luna OpenAI | 52.3 1428.5arena | $0.450 $0.200in$1.20out ×2 past 272K | 1.1M | frontier R T 3M |
| 17 Rank 17, tied with 8 others | Qwen3.8 27B Qwen | 52.0 1440.8arena | $1.05 $0.400in$3.00out | 1M | open R T 3M |
| 17 Rank 17, tied with 8 others | Gemini 3.5 Flash Google | 52.0 1482.6arena | $3.38 $1.50in$9.00out + reasoning billed | 1.0M | R T 5M |
| 17 Rank 17, tied with 8 others — on the value frontier | DeepSeek V4 Flash 0731 DeepSeek | 51.8 | $0.105 $0.080in$0.180out | 1.3M | frontier R T |
| 20 Rank 20, tied with 5 others | Gemini 3.6 Flash Google | 51.6 1476.5arena | $1.50 $0.750in$3.75out + reasoning billed | 1.0M | R T 5M |
| 26 Rank 26, tied with 1 others | Claude Sonnet 4.6 Anthropic | 48.4 1457.6arena | $6.00 $3.00in$15.00out | 1M | R T 3M |
| 26 Rank 26, tied with 2 others | Gemini 3.1 Pro Preview Google | 47.7 | $4.50 $2.00in$12.00out ×2 past 200K + reasoning billed | 1.0M | R T 5M |
| 27 Rank 27, tied with 3 others | Qwen3.7 Max Qwen | 46.7 1474.3arena | $2.21 $1.48in$4.42out | 1M | R T |
| 28 Rank 28, tied with 3 others | MiniMax M3 MiniMax | 45.4 1434.8arena | $0.525 $0.300in$1.20out | 1.0M | open R T 3M |
| 28 Rank 28, tied with 3 others | DeepSeek V4 Pro 0423 DeepSeek | 45.3 1450.9arena | $0.496 $0.397in$0.794out off-peak discount | 1.0M | open R T |
| 29 Rank 29, tied with 2 others | Kimi K2.6 Moonshot AI | 45.1 1455arena | $0.976 $0.541in$2.28out | 262K | R T 2M |
| 32 Rank 32, tied with 6 others | Kimi K2.7 Code Moonshot AI | 43.0 | $1.35 $0.670in$3.40out | 262K | open R T 2M |
| 32 Rank 32, tied with 6 others | MiMo-V2.5-Pro Xiaomi | 42.9 1465arena | $0.544 $0.435in$0.870out | 1.1M | R T |
| 32 Rank 32, tied with 9 others | Inkling Thinking Machines | 42.3 1439.2arena | $1.72 $0.950in$4.05out | 1.0M | R T 3M |
| 32 Rank 32, tied with 9 others | Hy3 preview Tencent | 42.2 | $0.285 $0.180in$0.600out | 262K | R T |
| 32 Rank 32, tied with 10 others — on the value frontier | DeepSeek V4 Flash 0423 DeepSeek | 42.1 1431.6arena | $0.061 $0.049in$0.098out off-peak discount | 1.0M | frontier open R T |
| 32 Rank 32, tied with 11 others | Nex-N2-Pro Nex Agi | 41.7 | $0.438 $0.250in$1.00out | 262K | R T 2M |
| 32 Rank 32, tied with 11 others — on the value frontier | Solar Pro 4 Upstage | 41.6 1376.2arena | $0.052 $0.030in$0.120out | 524K | frontier R T |
| 34 Rank 34, tied with 9 others | Inkling Small Thinking Machines | 41.2 1411.7arena | $0.637 $0.450in$1.20out | 1.0M | R T 3M |
| 34 Rank 34, tied with 10 others | GLM 5.1 Z.ai | 41.0 1464.1arena | $1.48 $0.966in$3.04out | 205K | R T |
| 34 Rank 34, tied with 10 others | GPT-5.4 Mini OpenAI | 40.9 1412.1arena | $1.69 $0.750in$4.50out | 400K | R T 3M |
| 36 Rank 36, tied with 9 others | Grok Build 0.1 xAI | 40.7 | $1.25 $1.00in$2.00out ×2 past 200K | 256K | R T 3M |
| 37 Rank 37, tied with 8 others | Qwen3.6 Plus Qwen | 40.5 1436.8arena | $0.731 $0.325in$1.95out ×4 past 256K | 1M | R T 3M |
| 40 Rank 40, tied with 7 others | GPT-5.4 Nano OpenAI | 39.7 1372.8arena | $0.463 $0.200in$1.25out | 400K | R T 3M |
| 42 Rank 42, tied with 6 others | Qwen3.7 Plus Qwen | 39.4 1456.2arena | $0.560 $0.320in$1.28out ×3 past 256K | 1M | R T 2M |
| 44 Rank 44, tied with 8 others | MiniMax M2.7 MiniMax | 38.9 1405.3arena | $0.420 $0.240in$0.960out | 205K | R T |
| 44 Rank 44, tied with 10 others | Nemotron 3 Ultra NVIDIA | 38.3 | $1.35 $0.600in$3.60out | 512K | open R T |
| 45 Rank 45, tied with 9 others | MiMo-V2.5 Xiaomi | 38.0 1427.3arena | $0.175 $0.140in$0.280out | 1.1M | R T 4M |
| 46 Rank 46, tied with 8 others | Grok 4.3 xAI | 37.9 1397.5arena | $1.56 $1.25in$2.50out ×2 past 200K | 1M | R T 3M |
| 46 Rank 46, tied with 8 others — on the value frontier | Ling-3.0-flash inclusionAI | 37.8 | $0.032 $0.021in$0.063out | 262K | frontier R T |
| 46 Rank 46, tied with 8 others | Qwen3.6 27B Qwen | 37.7 | $1.35 $0.600in$3.60out | 262K | R T 3M |
| 46 Rank 46, tied with 8 others | GPT-5.1 OpenAI | 37.5 1441.4arena | $3.44 $1.25in$10.00out | 400K | R T 3M |
| 47 Rank 47, tied with 8 others | Gemini 3.5 Flash Lite Google | 37.4 1436.5arena | $0.850 $0.300in$2.50out + reasoning billed | 1.0M | R T 5M |
| 47 Rank 47, tied with 8 others | Claude Sonnet 4.5 Anthropic | 37.4 | $6.00 $3.00in$15.00out ×2 past 200K | 1M | R T 3M |
| 53 Rank 53, tied with 3 others | Kimi K2.5 Moonshot AI | 36.0 1445.2arena | $0.900 $0.450in$2.25out | 262K | R T 2M |
| 55 Rank 55, tied with 4 others | GPT-5 OpenAI | 35.3 1405.6arena | $3.44 $1.25in$10.00out | 400K | R T 3M |
| 56 Rank 56, tied with 4 others | GLM 4.7 Z.ai | 34.5 1435.3arena | $0.738 $0.400in$1.75out | 205K | R T |
| 56 Rank 56, tied with 4 others | Qwen3.5 397B A17B Qwen | 34.3 1438.3arena | $0.877 $0.390in$2.34out | 262K | R T 3M |
| 56 Rank 56, tied with 5 others | LongCat 2.0 Meituan | 34.0 | $0.525 $0.300in$1.20out | 1.0M | R T |
| 57 Rank 57, tied with 4 others | KAT-Coder-Pro V2 Kwaipilot | 33.7 | $0.525 $0.300in$1.20out | 262K | T |
| 59 Rank 59, tied with 4 others | Qwen3.5-122B-A10B Qwen | 32.8 1417.9arena | $0.715 $0.260in$2.08out | 262K | R T 3M |
| 61 Rank 61, tied with 3 others | Qwen3.6 35B A3B Qwen | 32.1 | $0.355 $0.140in$1.00out | 262K | R T 3M |
| 61 Rank 61, tied with 4 others | DeepSeek V3.1 Terminus DeepSeek | 31.4 1419.6arena | $0.453 $0.270in$1.00out | 164K | R T |
| 62 Rank 62, tied with 6 others | Step 3.7 Flash StepFun | 30.9 | $0.438 $0.200in$1.15out | 262K | R T 3M |
| 63 Rank 63, tied with 6 others | Mistral Medium 3.5 Mistral | 30.4 1420.9arena | $3.00 $1.50in$7.50out | 262K | R T 3M |
| 64 Rank 64, tied with 5 others | Claude Haiku 4.5 Anthropic | 29.9 | $2.00 $1.00in$5.00out | 200K | R T 3M |
| 64 Rank 64, tied with 5 others | Claude Sonnet 4 Anthropic | 29.8 | $6.00 $3.00in$15.00out ×2 past 200K | 1M | R T 3M |
| 64 Rank 64, tied with 5 others | Gemma 4 31B Google | 29.7 1441.7arena | $0.160 $0.100in$0.340out | 262K | open R T 3M |
| 65 Rank 65, tied with 4 others | GLM 4.6 Z.ai | 29.3 1439.8arena | $0.875 $0.500in$2.00out | 205K | R T |
| 70 Rank 70, tied with 4 others | Gemma 4 26B A4B Google | 26.1 1434.6arena | $0.138 $0.070in$0.340out | 262K | R T 3M |
| 70 Rank 70, tied with 4 others | Gemini 2.5 Pro Google | 25.9 1457.3arena | $3.44 $1.25in$10.00out ×2 past 200K + reasoning billed | 1.0M | R T 5M |
| 70 Rank 70, tied with 4 others | GPT-5 Mini OpenAI | 25.8 1373.4arena | $0.688 $0.250in$2.00out | 400K | R T 3M |
| 70 Rank 70, tied with 5 others | Nemotron 3 Super NVIDIA | 25.7 | $0.164 $0.085in$0.400out | 1M | R T |
| 70 Rank 70, tied with 5 others | Gemini 3.1 Flash Lite Preview Google | 25.6 | $0.563 $0.250in$1.50out + reasoning billed | 1.0M | R T 5M |
| 73 Rank 73, tied with 4 others | Qwen3.5-35B-A3B Qwen | 24.3 1395.6arena | $0.500 $0.250in$1.25out | 262K | R T 3M |
| 75 Rank 75, tied with 3 others | gpt-oss-120b OpenAI | 24.1 1365.6arena | $0.070 $0.037in$0.170out | 131K | open R T |
| 75 Rank 75, tied with 3 others | Nemotron 3.5 Lightning NVIDIA | 23.6 | $0.110 $0.080in$0.200out | 262K | R T |
| 76 Rank 76, tied with 4 others | Command A Cohere | 22.8 | $4.38 $2.50in$10.00out | 256K | open |
| 78 Rank 78, tied with 3 others | Mercury 2 Inception | 21.9 1357.8arena | $0.375 $0.250in$0.750out | 128K | R T |
| 78 Rank 78, tied with 3 others | Qwen3.5-9B Qwen | 21.8 | $0.113 $0.100in$0.150out | 262K | R T 3M |
| 79 Rank 79, tied with 3 others | Qwen3 Coder Next Qwen | 21.3 | $0.290 $0.120in$0.800out | 262K | T |
| 81 Rank 81, tied with 4 others | Qwen3 235B A22B Thinking 2507 Qwen | 19.9 1413.8arena | $0.747 $0.230in$2.30out | 262K | R T |
| 82 Rank 82, tied with 3 others | Mistral Small 4 Mistral | 19.7 | $0.262 $0.150in$0.600out | 262K | open R T 2M |
| 82 Rank 82, tied with 3 others | Trinity Large Thinking Arcee | 18.7 | $0.378 $0.220in$0.850out | 262K | R T |
| 82 Rank 82, tied with 3 others | R1 DeepSeek | 18.6 1372.7arena | $1.15 $0.700in$2.50out | 64K | R T |
| 86 Rank 86, tied with 2 others | Qwen3 Next 80B A3B Thinking Qwen | 16.9 | $0.412 $0.150in$1.20out | 262K | R T |
| 86 Rank 86, tied with 10 others | Mistral Large 3 2512 Mistral | 15.9 | $0.750 $0.500in$1.50out | 262K | T 3M |
| 86 Rank 86, tied with 10 others | o3 Mini High OpenAI | 15.7 | $1.93 $1.10in$4.40out | 200K | R T 2M |
| 87 Rank 87, tied with 9 others | gpt-oss-20b OpenAI | 15.2 1287.8arena | $0.055 $0.030in$0.130out | 131K | R T |
| 87 Rank 87, tied with 9 others | DeepSeek V3 0324 DeepSeek | 15.2 1375arena | $0.438 $0.250in$1.00out | 164K | T |
| 87 Rank 87, tied with 9 others | GPT-4.1 Mini OpenAI | 14.8 | $0.700 $0.400in$1.60out | 1.0M | T 3M |
| 87 Rank 87, tied with 9 others | Mistral Medium 3.1 Mistral | 14.7 | $0.800 $0.400in$2.00out | 131K | T 3M |
| 87 Rank 87, tied with 9 others | Qwen3 30B A3B Thinking 2507 Qwen | 14.6 | $0.750 $0.200in$2.40out | 82K | R T |
| 87 Rank 87, tied with 9 others | Nemotron 3 Nano 30B A3B NVIDIA | 14.5 | $0.088 $0.050in$0.200out | 262K | R T |
| 87 Rank 87, tied with 9 others | Solar Pro 3 Upstage | 14.5 | $0.262 $0.150in$0.600out | 131K | R T |
| 87 Rank 87, tied with 9 others | Llama 4 Maverick Meta | 14.5 | $0.350 $0.200in$0.800out | 1.0M | open T 2M |
| 97 Rank 97, tied with 3 others | Qwen3 32B Qwen | 11.4 1340.1arena | $0.130 $0.080in$0.280out | 131K | R T |
| 97 Rank 97, tied with 3 others | Ministral 3 14B 2512 Mistral | 11.2 | $0.200 $0.200in$0.200out | 262K | T 2M |
| 97 Rank 97, tied with 5 others | Qwen3 14B Qwen | 10.4 | $0.150 $0.120in$0.240out | 131K | R T |
| 97 Rank 97, tied with 5 others | Llama 4 Scout Meta | 10.3 | $0.150 $0.100in$0.300out | 1.3M | open T 2M |
| 99 Rank 99, tied with 4 others | GPT-4.1 Nano OpenAI | 9.6 | $0.175 $0.100in$0.400out | 1.0M | T 3M |
| 99 Rank 99, tied with 4 others | Ministral 3 8B 2512 Mistral | 9.0 | $0.150 $0.150in$0.150out | 262K | T 2M |
| 101 Rank 101, tied with 4 others | Qwen3 8B Qwen | 8.3 | $0.202 $0.117in$0.455out | 131K | R T |
| 103 Rank 103, tied with 2 others | Gemma 3 27B Google | 7.4 1358.3arena | $0.172 $0.080in$0.450out | 262K | T 2M |
| 103 Rank 103, tied with 2 others | Ministral 3 3B 2512 Mistral | 7.1 | $0.100 $0.100in$0.100out | 131K | T 2M |
| 106 Rank 106 | Gemma 3 12B Google | 5.5 1334.2arena | $0.075 $0.050in$0.150out | 131K | T 2M |
| 221 models have no independent quality score. They are listed by price. Not benchmarked is not the same as benchmarked badly. | |||||
| — | Ox Alpha Stealth | unrated | free freeinfreeout | 1.0M | R T 3M |
| — | Lyria 3 Pro Preview Google | unrated | free freeinfreeout | 1.0M | 2M |
| — | Lyria 3 Clip Preview Google | unrated | free freeinfreeout | 1.0M | 2M |
| — | Free Models Router OpenRouter | unrated | free freeinfreeout | 200K | T 2M |
| — | Mistral Nemo Mistral | unrated | $0.022 $0.019in$0.030out | 131K | T |
| — | Granite 4.0 Micro IBM | unrated | $0.041 $0.017in$0.112out | 131K | |
| — | Llama 3 8B Lunaris Sao10k | unrated | $0.042 $0.040in$0.050out | 8K | |
| — | Nex-N2-Mini Nex Agi | unrated | $0.044 $0.025in$0.100out | 262K | R T 2M |
| — | Qwen3.7 Flash Qwen | unrated | $0.055 $0.030in$0.130out ×3.3 past 32K+ | 1M | R T 3M |
| — | Mistral Small 3 Mistral | unrated | $0.058 $0.050in$0.080out | 33K | |
| — | Llama 3.1 8B Instruct Meta | unrated | $0.058 $0.050in$0.080out | 131K | T |
| — | MythoMax 13B Gryphe | unrated | $0.060 $0.060in$0.060out | 8K | |
| — | Nova Micro 1.0 Amazon | unrated | $0.061 $0.035in$0.140out | 128K | T |
| — | Granite 4.1 8B IBM | unrated | $0.063 $0.050in$0.100out | 131K | open T |
| — | Gemma 3 4B Google | unrated | $0.063 $0.050in$0.100out | 131K | 2M |
| — | Command R7B (12-2024) Cohere | unrated | $0.066 $0.037in$0.150out | 128K | |
| — | Llama 3.2 1B Instruct Meta | unrated | $0.071 $0.027in$0.201out | 60K | |
| — | Laguna XS 2.1 Poolside | unrated | $0.075 $0.060in$0.120out | 262K | R T |
| — | Gemma 3n 4B Google | unrated | $0.075 $0.060in$0.120out | 33K | |
| — | Hy-MT2-1.8B Tencent | unrated | $0.077 $0.044in$0.177out | 8K | |
| — | Qwen3 30B A3B Instruct 2507 Qwen | unrated | $0.084 $0.048in$0.193out | 262K | T |
| — | Phi 4 Microsoft | unrated | $0.088 $0.070in$0.140out | 16K | open |
| — | Reka Edge Reka | unrated | $0.100 $0.100in$0.100out | 16K | R T 3M |
| — | Nova Lite 1.0 Amazon | unrated | $0.105 $0.060in$0.240out | 300K | T 2M |
| — | Mistral Small 3.2 24B Mistral | unrated | $0.106 $0.075in$0.200out | 131K | T 2M |
| — | Ministral 8B Mistral | unrated | $0.110 $0.110in$0.110out | 128K | |
| — | Laguna S 2.1 Poolside | unrated | $0.113 $0.090in$0.180out | 1.0M | R T |
| — | Qwen3.5-Flash Qwen | unrated | $0.114 $0.065in$0.260out | 1M | R T 3M |
| — | Llama 3.2 3B Instruct Meta | unrated | $0.120 $0.050in$0.330out | 131K | |
| — | Qwen3 Coder 30B A3B Instruct Qwen | unrated | $0.123 $0.070in$0.280out | 262K | T |
| — | Muse Spark 1.2 Contributor Meta | unrated | $0.125 $0.100in$0.200out | 1.0M | R T 5M |
| — | UI-TARS 7B ByteDance | unrated | $0.125 $0.100in$0.200out | 128K | 2M |
| — | Reka Flash 3 Reka | unrated | $0.125 $0.100in$0.200out | 66K | R |
| — | Qwen2.5 7B Instruct Qwen | unrated | $0.125 $0.100in$0.200out | 33K | T |
| — | Hy-MT2-30B-A3B Tencent | unrated | $0.129 $0.074in$0.295out | 8K | |
| — | Hy-MT2-7B Tencent | unrated | $0.129 $0.074in$0.295out | 8K | |
| — | Seed 1.6 Flash ByteDance | unrated | $0.131 $0.075in$0.300out ×1.3 past 128K | 262K | R T 3M |
| — | gpt-oss-safeguard-20b OpenAI | unrated | $0.131 $0.075in$0.300out | 131K | R T |
| — | GPT-5 Nano OpenAI | unrated | $0.138 $0.050in$0.400out | 400K | R T 3M |
| — | GLM 4.7 Flash Z.ai | unrated | $0.145 $0.060in$0.400out | 203K | R T |