Best models first. Then what they cost.
125 of 134 rated, priced models are beaten on quality and undercut on price by something else on this page. 9 are not.
Measured on
General · Balanced · Short · 134 ranked · 193 unrated · 64 distinct ranks
Advanced filters
Best for
Workload
Prompt size
Refine
134 ranked· 193 unrated
How these ranks work
The tinted bands are the ranks the measurement can actually support. 134 models occupy 64 genuinely distinct ranks. The integer is secondary. A dotted marker means a statistical tie. Computed from each model’s own published confidence interval.
| # | Beaten by | Notes | ||||
|---|---|---|---|---|---|---|
| 1Rank 1, tied with 2 others— on the value frontier | Claude Opus 5Anthropic | 1504.2 |
$10.00$5.00in$25.00out |
— | 1M | frontierRReasoningTTools3M3 modalities |
| 1Rank 1, tied with 2 others | Claude Opus 4.6Anthropic | 1502.7 |
$10.00$5.00in$25.00out |
Beaten byClaude Opus 5 | 1M | RReasoningTTools3M3 modalities |
| 1Rank 1, tied with 4 others | Claude Fable 5Anthropic | 1494.7 |
$20.00$10.00in$50.00out |
Beaten byClaude Opus 5 | 1M | RReasoningTTools3M3 modalities |
| 3Rank 3, tied with 8 others— on the value frontier | Gemini 3.7 FlashGoogle | 1490.2 |
$0.750$0.375in$1.88out+ reasoning billed |
— | 1.0M | frontierRReasoningTTools5M5 modalities |
| 3Rank 3, tied with 5 others | Claude Opus 4.7Anthropic | 1490.1 |
$10.00$5.00in$25.00out |
Beaten byClaude Opus 5 | 1M | RReasoningTTools3M3 modalities |
| 4Rank 4, tied with 8 others | Gemini 3.5 FlashGoogle | 1482.6 |
$3.38$1.50in$9.00out+ reasoning billed |
Beaten byGemini 3.7 Flash | 1.0M | RReasoningTTools5M5 modalities |
| 4Rank 4, tied with 8 others | Qwen3.8 MaxQwen | 1481.9 |
$3.00$2.00in$6.00out |
Beaten byGemini 3.7 Flashmax output | 1M | RReasoningTTools3M3 modalities |
| 5Rank 5, tied with 7 others | Gemini 3.1 Pro PreviewGoogle | 1479.5 |
$4.50$2.00in$12.00out×2 past 200K+ reasoning billed |
Beaten byGemini 3.7 Flash | 1.0M | RReasoningTTools5M5 modalities |
| 5Rank 5, tied with 9 others | Muse Spark 1.1Meta | 1478.3 |
$2.00$1.25in$4.25out |
Beaten byGemini 3.7 Flashmax output | 1.0M | RReasoningTTools5M5 modalities |
| 4Rank 4, tied with 13 others | GLM 5.3Z.ai | 1476.9 |
$2.15$1.40in$4.40out |
Beaten byMuse Spark 1.1 | 1.0M | RReasoningTTools |
| 6Rank 6, tied with 8 others | Gemini 3.6 FlashGoogle | 1476.5 |
$1.50$0.750in$3.75out+ reasoning billed |
Beaten byGemini 3.7 Flash | 1.0M | RReasoningTTools5M5 modalities |
| 5Rank 5, tied with 9 others | Kimi K3Moonshot AI | 1476.3 |
$6.00$3.00in$15.00outlike-for-like $2.85 |
Beaten byMuse Spark 1.1 | 1.0M | openRReasoningTTools3M3 modalities |
| 9Rank 9, tied with 8 others | GPT-5.5OpenAI | 1470.9 |
$11.25$5.00in$30.00out×2 past 272K×2.2 |
Beaten byClaude Opus 5context | 1.1M | RReasoningTTools3M3 modalities |
| 9Rank 9, tied with 8 others | GPT-5.4OpenAI | 1469.6 |
$5.63$2.50in$15.00out×2 past 272K×2.2 |
Beaten byGemini 3.7 Flashcontext, max output | 1.1M | RReasoningTTools3M3 modalities |
| 12Rank 12, tied with 8 others | GLM 5.2Z.ai | 1465.4 |
$1.83$1.19in$3.74out×2.8 |
Beaten byGemini 3.7 Flashmax output | 1.0M | openRReasoningTTools |
| 12Rank 12, tied with 7 others— on the value frontier | MiMo-V2.5-ProXiaomi | 1465.0 |
$0.544$0.435in$0.870out×3.28 |
— | 1.1M | frontierRReasoningTTools |
| 12Rank 12, tied with 8 others | GLM 5.1Z.ai | 1464.1 |
$1.94$1.26in$3.96out×1.69 |
Beaten byGLM 5.2 | 205K | RReasoningTTools |
| 15Rank 15, tied with 9 others | Claude Opus 4.8Anthropic | 1460.9 |
$10.00$5.00in$25.00out |
Beaten byClaude Opus 5 | 1M | RReasoningTTools3M3 modalities |
| 15Rank 15, tied with 10 others | Claude Sonnet 4.6Anthropic | 1457.6 |
$6.00$3.00in$15.00out |
Beaten byMuse Spark 1.1 | 1M | RReasoningTTools3M3 modalities |
| 18Rank 18, tied with 6 others | Gemini 2.5 ProGoogle | 1457.3 |
$3.44$1.25in$10.00out×2 past 200K+ reasoning billed |
Beaten byGemini 3.7 Flash | 1.0M | RReasoningTTools5M5 modalities |
| 16Rank 16, tied with 11 others | Qwen3.7 PlusQwen | 1456.2 |
$0.560$0.320in$1.28out×3 past 256K |
Beaten byMiMo-V2.5-Proimage | 1M | RReasoningTTools2M2 modalities |
| 18Rank 18, tied with 11 others | Kimi K2.6Moonshot AI | 1455.0 |
$1.71$0.950in$4.00outlike-for-like $0.700 |
Beaten byGemini 3.7 Flashmax output | 262K | RReasoningTTools2M2 modalities |
| 18Rank 18, tied with 15 others | GPT-5.6 SolOpenAI | 1454.2 |
$4.00$2.00in$10.00out×2 past 272K×5.5 |
Beaten byGemini 3.7 Flashcontext, max output | 1.1M | RReasoningTTools3M3 modalities |
| 18Rank 18, tied with 17 others | Grok 4.5xAI | 1452.3 |
$3.00$2.00in$6.00out×2 past 200K |
Beaten byMuse Spark 1.1 | 500K | RReasoningTTools3M3 modalities |
| 21Rank 21, tied with 14 others | Claude Opus 4.5Anthropic | 1449.8 |
$10.00$5.00in$25.00out |
Beaten byClaude Opus 5 | 200K | RReasoningTTools3M3 modalities |
| 20Rank 20, tied with 26 others | Qwen3.6 Max PreviewQwen | 1446.3 |
$2.31$1.03in$6.16out×1.5 past 128K |
Beaten byGemini 3.7 Flash | 262K | RReasoningTTools |
| 22Rank 22, tied with 23 others | GPT-5.6 TerraOpenAI | 1446.0 |
$4.50$2.00in$12.00out×2 past 272K×2.2 |
Beaten byGPT-5.6 Sol | 1.1M | RReasoningTTools3M3 modalities |
| 24Rank 24, tied with 17 others | Kimi K2.5Moonshot AI | 1445.2 |
$0.900$0.450in$2.25outlike-for-like $0.540 |
Beaten byGemini 3.7 Flashmax output | 262K | RReasoningTTools2M2 modalities |
| 23Rank 23, tied with 21 others | GLM 5Z.ai | 1445.2 |
$0.930$0.600in$1.92out×1.67 |
Beaten byMiMo-V2.5-Pro | 205K | RReasoningTTools |
| 21Rank 21, tied with 28 others | Grok 4.6xAI | 1443.7 |
$3.00$2.00in$6.00out×2 past 200K |
Beaten byMuse Spark 1.1 | 500K | RReasoningTTools3M3 modalities |
| 24Rank 24, tied with 22 others | Claude Sonnet 5Anthropic | 1442.3 |
$4.00$2.00in$10.00out |
Beaten byMuse Spark 1.1 | 1M | RReasoningTTools3M3 modalities |
| 23Rank 23, tied with 25 others— on the value frontier | Gemma 4 31BGoogle | 1441.7 |
$0.153$0.090in$0.340out×12.38 |
— | 262K | frontieropenRReasoningTTools3M3 modalities |
| 26Rank 26, tied with 20 others | GPT-5.1OpenAI | 1441.4 |
$3.44$1.25in$10.00out×2 |
Beaten byMuse Spark 1.1 | 400K | RReasoningTTools3M3 modalities |
| 23Rank 23, tied with 27 others— on the value frontier | Hy3Tencent | 1441.2 |
$0.144$0.083in$0.330outoff-peak discount×1.59 |
— | 262K | frontierRReasoningTTools |
| 22Rank 22, tied with 30 others | Qwen3.8 27BQwen | 1440.8 |
$0.956$0.425in$2.55outlike-for-like $0.400 |
Beaten byGemini 3.7 Flashmax output | 1M | openRReasoningTTools3M3 modalities |
| 26Rank 26, tied with 20 others | GLM 4.6Z.ai | 1439.8 |
$0.875$0.500in$2.00outlike-for-like $0.550 |
Beaten byMiMo-V2.5-Pro | 205K | RReasoningTTools |
| 23Rank 23, tied with 30 others | DeepSeek V4 Pro 0423DeepSeek | 1439.2 |
$1.09$0.870in$1.74outoff-peak discount×3.66 |
Beaten byGemini 3.7 Flashmax output | 1.0M | openRReasoningTTools |
| 26Rank 26, tied with 22 others | InklingThinking Machines | 1439.2 |
$1.72$0.950in$4.05out |
Beaten byGemini 3.7 Flashmax output | 1.0M | RReasoningTTools3M3 modalities |
| 27Rank 27, tied with 20 others | Qwen3.5 397B A17BQwen | 1438.3 |
$0.877$0.390in$2.34outlike-for-like $0.450 |
Beaten byGemini 3.7 Flash | 262K | RReasoningTTools3M3 modalities |
| 28Rank 28, tied with 19 others | Claude Sonnet 4.5Anthropic | 1438.0 |
$6.00$3.00in$15.00out×2 past 200K |
Beaten byGemini 3.7 Flash | 1M | RReasoningTTools3M3 modalities |
| 27Rank 27, tied with 23 others | Qwen3.6 PlusQwen | 1436.8 |
$0.731$0.325in$1.95out×4 past 256K |
Beaten byMiMo-V2.5-Proimage, video | 1M | RReasoningTTools3M3 modalities |
| 26Rank 26, tied with 26 others | GLM 5V TurboZ.ai | 1436.8 |
$1.90$1.20in$4.00out |
Beaten byQwen3.8 27B | 203K | RReasoningTTools3M3 modalities |
| 26Rank 26, tied with 25 others | Gemini 3.5 Flash LiteGoogle | 1436.5 |
$0.850$0.300in$2.50out+ reasoning billed |
Beaten byGemini 3.7 Flash | 1.0M | RReasoningTTools5M5 modalities |
| 27Rank 27, tied with 26 others | GLM 4.7Z.ai | 1435.3 |
$0.738$0.400in$1.75outlike-for-like $0.480 |
Beaten byMiMo-V2.5-Pro | 205K | RReasoningTTools |
| 29Rank 29, tied with 23 others | MiniMax M3MiniMax | 1434.8 |
$0.525$0.300in$1.20outlike-for-like $0.240 |
Beaten byGemma 4 31Bcontext, max output | 1.0M | openRReasoningTTools3M3 modalities |
| 26Rank 26, tied with 30 others— on the value frontier | Gemma 4 26B A4B Google | 1434.6 |
$0.138$0.070in$0.340outlike-for-like $0.100 |
— | 262K | frontierRReasoningTTools3M3 modalities |
| 33Rank 33, tied with 21 others— on the value frontier | DeepSeek V4 Flash 0423DeepSeek | 1431.6 |
$0.101$0.081in$0.162outoff-peak discount×7.86 |
— | 1.0M | frontieropenRReasoningTTools |
| 35Rank 35, tied with 22 others | GLM 4.5Z.ai | 1429.4 |
$1.00$0.600in$2.20out |
Beaten byMiMo-V2.5-Pro | 131K | RReasoningTTools |
| 37Rank 37, tied with 21 others | GPT-5.6 LunaOpenAI | 1428.5 |
$0.450$0.200in$1.20out×2 past 272K×2.2 |
Beaten byGemma 4 31Bcontext, max output, file | 1.1M | RReasoningTTools3M3 modalities |
| 38Rank 38, tied with 20 others | R1 0528DeepSeek | 1427.9 |
$0.912$0.500in$2.15out |
Beaten byGemini 3.7 Flash | 164K | RReasoningTTools |
| 40Rank 40, tied with 18 others | MiMo-V2.5Xiaomi | 1427.3 |
$0.175$0.140in$0.280outlike-for-like $0.400 |
Beaten byGemma 4 31Bcontext, max output, audio | 1.1M | RReasoningTTools4M4 modalities |
| 46Rank 46, tied with 14 others | DeepSeek V3.2DeepSeek | 1424.6 |
$0.290$0.260in$0.380out×14.37 |
Beaten byDeepSeek V4 Flash 0423 | 164K | openRReasoningTTools |
| 41Rank 41, tied with 26 others | DeepSeek V3.2 ExpDeepSeek | 1424.4 |
$0.305$0.270in$0.410out |
Beaten byHy3 | 164K | RReasoningTTools |
| 47Rank 47, tied with 21 others | Qwen3 VL 235B A22B InstructQwen | 1420.9 |
$0.632$0.210in$1.90outlike-for-like $0.300 |
Beaten byQwen3.7 Plus | 262K | TTools2M2 modalities |
| 47Rank 47, tied with 21 others | Mistral Medium 3.5Mistral | 1420.9 |
$3.00$1.50in$7.50out |
Beaten byMuse Spark 1.1 | 262K | RReasoningTTools3M3 modalities |
| 44Rank 44, tied with 27 others | DeepSeek V3.1 TerminusDeepSeek | 1419.6 |
$0.453$0.270in$1.00out |
Beaten byHy3 | 164K | RReasoningTTools |
| 52Rank 52, tied with 15 others | Qwen3 235B A22B Instruct 2507Qwen | 1419.3 |
$0.205$0.090in$0.550out×2.51 |
Beaten byGemma 4 31B | 262K | TTools |
| 48Rank 48, tied with 20 others | DeepSeek V3.1DeepSeek | 1419.1 |
$0.825$0.550in$1.65outlike-for-like $0.270 |
Beaten byMiniMax M3 | 164K | RReasoningTTools |
| 49Rank 49, tied with 19 others | Qwen3 Next 80B A3B InstructQwen | 1418.6 |
$0.350$0.100in$1.10outlike-for-like $0.150 |
Beaten byDeepSeek V4 Flash 0423 | 262K | TTools |
| 52Rank 52, tied with 16 others | Qwen3.5-122B-A10BQwen | 1417.9 |
$0.715$0.260in$2.08outlike-for-like $0.400 |
Beaten byMiniMax M3 | 262K | RReasoningTTools3M3 modalities |
| 53Rank 53, tied with 15 others | Claude Opus 4.1Anthropic | 1417.6 |
$30.00$15.00in$75.00out |
Beaten byClaude Opus 5 | 200K | RReasoningTTools3M3 modalities |
| 53Rank 53, tied with 15 others | Gemini 2.5 FlashGoogle | 1417.3 |
$0.850$0.300in$2.50out+ reasoning billed |
Beaten byGemini 3.7 Flash | 1.0M | RReasoningTTools5M5 modalities |
| 53Rank 53, tied with 15 others | GPT-5.2OpenAI | 1416.5 |
$4.81$1.75in$14.00out×2 |
Beaten byMuse Spark 1.1 | 400K | RReasoningTTools3M3 modalities |
| 53Rank 53, tied with 17 others | Gemini 3.1 Flash Lite PreviewGoogle | 1414.8 |
$0.563$0.250in$1.50out+ reasoning billed |
Beaten byMiMo-V2.5-Proimage, video, file, audio | 1.0M | RReasoningTTools5M5 modalities |
| 53Rank 53, tied with 21 others | Qwen3 235B A22B Thinking 2507Qwen | 1413.8 |
$0.747$0.230in$2.30outlike-for-like $0.300 |
Beaten byMiMo-V2.5-Pro | 131K | RReasoningTTools |
| 53Rank 53, tied with 21 others | Qwen3 MaxQwen | 1412.7 |
$1.56$0.780in$3.90out×2 past 32K+ |
Beaten byGemini 3.7 Flash | 262K | RReasoningTTools |
| 55Rank 55, tied with 17 others | GPT-5.4 MiniOpenAI | 1412.1 |
$1.69$0.750in$4.50out×2 |
Beaten byGPT-5.6 Luna | 400K | RReasoningTTools3M3 modalities |
| 53Rank 53, tied with 21 others | Inkling SmallThinking Machines | 1411.7 |
$0.637$0.450in$1.20out |
Beaten byMiMo-V2.5-Promax output, image, audio | 1.0M | RReasoningTTools3M3 modalities |
| 63Rank 63, tied with 11 others | o3OpenAI | 1409.2 |
$3.50$2.00in$8.00out |
Beaten byMuse Spark 1.1 | 200K | RReasoningTTools3M3 modalities |
| 63Rank 63, tied with 11 others | Qwen3.5-27BQwen | 1407.9 |
$0.536$0.195in$1.56outlike-for-like $0.300 |
Beaten byMiniMax M3 | 262K | RReasoningTTools3M3 modalities |
| 64Rank 64, tied with 12 others | GPT-5OpenAI | 1405.6 |
$3.44$1.25in$10.00out |
Beaten byMuse Spark 1.1 | 400K | RReasoningTTools3M3 modalities |
| 65Rank 65, tied with 9 others | MiniMax M2.7MiniMax | 1405.3 |
$0.525$0.300in$1.20out×2.5 |
Beaten byMiniMax M3 | 205K | RReasoningTTools |
| 66Rank 66, tied with 10 others | Step 3.5 FlashStepFun | 1403.8 |
$0.150$0.100in$0.300out |
Beaten byHy3 | 262K | RReasoningTTools |
| 66Rank 66, tied with 12 others | Qwen3 VL 235B A22B ThinkingQwen | 1400.6 |
$1.30$0.400in$4.00outlike-for-like $0.980 |
Beaten byGemini 3.7 Flash | 131K | RReasoningTTools2M2 modalities |
| 72Rank 72, tied with 6 others | Qwen3.5-FlashQwen | 1397.6 |
$0.114$0.065in$0.260out |
Beaten byDeepSeek V4 Flash 0423image, video | 1M | RReasoningTTools3M3 modalities |
| 72Rank 72, tied with 6 others | Grok 4.3xAI | 1397.5 |
$1.56$1.25in$2.50out×2 past 200K |
Beaten byGemini 3.7 Flashmax output | 1M | RReasoningTTools3M3 modalities |
| 74Rank 74, tied with 4 others | Qwen3.5-35B-A3BQwen | 1395.6 |
$0.500$0.250in$1.25out×2.23 |
Beaten byGemma 4 31Bmax output | 262K | RReasoningTTools3M3 modalities |
| 74Rank 74, tied with 4 others | Claude Haiku 4.5Anthropic | 1394.9 |
$2.00$1.00in$5.00out |
Beaten byGemini 3.7 Flash | 200K | RReasoningTTools3M3 modalities |
| 79Rank 79, tied with 4 others— on the value frontier | Qwen3 30B A3B Instruct 2507Qwen | 1384.3 |
$0.084$0.048in$0.193outlike-for-like $0.100 |
— | 262K | frontierTTools |
| 79Rank 79, tied with 5 others | GLM 4.5 AirZ.ai | 1382.8 |
$0.310$0.130in$0.850out×1.54 |
Beaten byHy3 | 131K | RReasoningTTools |
| 79Rank 79, tied with 5 others | GPT-4.1OpenAI | 1382.3 |
$3.50$2.00in$8.00out |
Beaten byGemini 3.7 Flash | 1.0M | TTools3M3 modalities |
| 79Rank 79, tied with 15 others— on the value frontier | Solar Pro 4Upstage | 1376.2 |
$0.052$0.030in$0.120out |
— | 524K | frontierRReasoningTTools |
| 80Rank 80, tied with 8 others | DeepSeek V3 0324DeepSeek | 1375.0 |
$0.438$0.250in$1.00outlike-for-like $0.500 |
Beaten byDeepSeek V4 Flash 0423 | 164K | TTools |
| 79Rank 79, tied with 15 others | GLM 4.6VZ.ai | 1374.7 |
$0.450$0.300in$0.900out |
Beaten byMiMo-V2.5 | 131K | RReasoningTTools3M3 modalities |
| 82Rank 82, tied with 9 others | GPT-5 MiniOpenAI | 1373.4 |
$0.688$0.250in$2.00out×2 |
Beaten byGPT-5.6 Luna | 400K | RReasoningTTools3M3 modalities |
| 82Rank 82, tied with 9 others | GPT-5.4 NanoOpenAI | 1372.8 |
$0.463$0.200in$1.25out×2 |
Beaten byGPT-5.6 Luna | 400K | RReasoningTTools3M3 modalities |
| 82Rank 82, tied with 12 others | R1DeepSeek | 1372.7 |
$1.15$0.700in$2.50out |
Beaten byGemini 3.7 Flash | 64K | RReasoningTTools |
| 82Rank 82, tied with 14 others | Qwen3 Next 80B A3B ThinkingQwen | 1367.5 |
$0.412$0.150in$1.20out |
Beaten byHy3 | 262K | RReasoningTTools |
| 83Rank 83, tied with 13 others | Qwen3 235B A22BQwen | 1366.0 |
$0.796$0.455in$1.82out |
Beaten byGemini 3.7 Flash | 131K | RReasoningTTools |
| 83Rank 83, tied with 13 others | o1OpenAI | 1365.9 |
$26.25$15.00in$60.00out |
Beaten byClaude Opus 5 | 200K | RReasoningTTools3M3 modalities |
| 83Rank 83, tied with 13 others | gpt-oss-120bOpenAI | 1365.6 |
$0.070$0.037in$0.170outlike-for-like $0.037 |
Beaten bySolar Pro 4 | 131K | openRReasoningTTools |
| 85Rank 85, tied with 11 others | Claude Opus 4Anthropic | 1364.3 |
$30.00$15.00in$75.00out |
Beaten byClaude Opus 5 | 200K | RReasoningTTools3M3 modalities |
| 85Rank 85, tied with 13 others | Nova 2 LiteAmazon | 1362.2 |
$0.850$0.300in$2.50out |
Beaten byGemini 3.7 Flash | 1M | RReasoningTTools4M4 modalities |
| 88Rank 88, tied with 10 others | MiniMax M2.5MiniMax | 1359.0 |
$0.473$0.270in$1.08outlike-for-like $0.295 |
Beaten byHy3 | 205K | RReasoningTTools |
| 88Rank 88, tied with 10 others | Gemma 3 27BGoogle | 1358.3 |
$0.172$0.080in$0.450outlike-for-like $0.119 |
Beaten byGemma 4 31Bmax output | 262K | TTools2M2 modalities |
| 85Rank 85, tied with 15 others | Mercury 2Inception | 1357.8 |
$0.375$0.250in$0.750out |
Beaten byHy3 | 128K | RReasoningTTools |
| 93Rank 93, tied with 7 others | GLM 4.7 FlashZ.ai | 1352.9 |
$0.145$0.060in$0.400out |
Beaten byHy3 | 203K | RReasoningTTools |
| 93Rank 93, tied with 7 others | o4 MiniOpenAI | 1352.5 |
$1.93$1.10in$4.40out |
Beaten byGPT-5.6 Luna | 200K | RReasoningTTools3M3 modalities |
| 96Rank 96, tied with 12 others | MiniMax M2MiniMax | 1342.1 |
$0.446$0.255in$1.02out |
Beaten byDeepSeek V4 Flash 0423 | 205K | RReasoningTTools |
| 99Rank 99, tied with 9 others | Trinity Large ThinkingArcee | 1341.9 |
$0.378$0.220in$0.850out |
Beaten byDeepSeek V4 Flash 0423 | 262K | RReasoningTTools |
| 99Rank 99, tied with 8 others | MiniMax M1MiniMax | 1341.9 |
$0.963$0.550in$2.20outlike-for-like $0.550 |
Beaten byGemini 3.7 Flash | 1M | RReasoningTTools |
| 99Rank 99, tied with 9 others | GPT-4.1 MiniOpenAI | 1340.5 |
$0.700$0.400in$1.60out |
Beaten byGPT-5.6 Luna | 1.0M | TTools3M3 modalities |
| 96Rank 96, tied with 12 others | Qwen3 32BQwen | 1340.1 |
$0.130$0.080in$0.280out×3.62 |
Beaten byDeepSeek V4 Flash 0423 | 131K | RReasoningTTools |
| 99Rank 99, tied with 9 others | Claude Sonnet 4Anthropic | 1338.2 |
$6.00$3.00in$15.00out×2 past 200K |
Beaten byGemini 3.7 Flash | 1M | RReasoningTTools3M3 modalities |
| 99Rank 99, tied with 9 others | o3 Mini HighOpenAI | 1336.6 |
$1.93$1.10in$4.40out |
Beaten byGPT-5.6 Luna | 200K | RReasoningTTools2M2 modalities |
| 99Rank 99, tied with 10 others | Gemma 3 12BGoogle | 1334.2 |
$0.075$0.050in$0.150out |
Beaten bySolar Pro 4image | 131K | TTools2M2 modalities |
| 99Rank 99, tied with 10 others | GLM 4.5VZ.ai | 1333.6 |
$0.900$0.600in$1.80out |
Beaten byGemini 3.7 Flash | 66K | RReasoningTTools2M2 modalities |
| 100Rank 100, tied with 8 others | DeepSeek V3DeepSeek | 1332.6 |
$0.450$0.257in$1.03outlike-for-like $0.400 |
Beaten byGemma 4 31B | 164K | TTools |
| 107Rank 107, tied with 4 others | GPT-5 NanoOpenAI | 1320.3 |
$0.138$0.050in$0.400out×2 |
Beaten byGemma 4 26B A4B context, max output, file | 400K | RReasoningTTools3M3 modalities |
| 109Rank 109, tied with 2 others | o3 MiniOpenAI | 1319.2 |
$1.93$1.10in$4.40out |
Beaten byGPT-5.6 Luna | 200K | RReasoningTTools2M2 modalities |
| 109Rank 109, tied with 2 others | Qwen3 30B A3BQwen | 1316.9 |
$0.215$0.120in$0.500out |
Beaten byGemma 4 31B | 131K | RReasoningTTools |
| 112Rank 112, tied with 3 others | GPT-4o (2024-05-13)OpenAI | 1300.6 |
$7.50$5.00in$15.00out |
Beaten byGemini 3.7 Flash | 128K | TTools3M3 modalities |
| 112Rank 112, tied with 5 others | Olmo 3 32B ThinkAllen AI | 1298.5 |
$0.237$0.150in$0.500out |
Beaten byHy3 | 66K | openRReasoning |
| 112Rank 112, tied with 7 others | Granite 4.1 8BIBM | 1291.6 |
$0.063$0.050in$0.100out |
Beaten bySolar Pro 4 | 131K | openTTools |
| 112Rank 112, tied with 7 others | Gemma 3 4BGoogle | 1290.8 |
$0.063$0.050in$0.100out |
Beaten bySolar Pro 4image | 131K | 2M2 modalities |
| 113Rank 113, tied with 6 others | gpt-oss-20bOpenAI | 1287.8 |
$0.055$0.030in$0.130out×2.5 |
Beaten bySolar Pro 4 | 131K | RReasoningTTools |
| 114Rank 114, tied with 5 others | GPT-4o-mini (2024-07-18)OpenAI | 1286.6 |
$0.262$0.150in$0.600out |
Beaten byGPT-5 Nano | 128K | TTools3M3 modalities |
| 113Rank 113, tied with 7 others | GPT-4.1 NanoOpenAI | 1284.8 |
$0.175$0.100in$0.400out |
Beaten byGemma 4 31Bcontext, max output, file | 1.0M | TTools3M3 modalities |
| 114Rank 114, tied with 5 others | GPT-4o (2024-08-06)OpenAI | 1282.7 |
$4.38$2.50in$10.00out |
Beaten byGemini 3.7 Flash | 128K | TTools3M3 modalities |
| 119Rank 119, tied with 3 others | Llama 3.3 70B InstructMeta | 1274.9 |
$0.710$0.710in$0.710outlike-for-like $0.135 |
Beaten byMiMo-V2.5-Pro | 131K | TTools |
| 120Rank 120, tied with 3 others | GPT-4 TurboOpenAI | 1271.8 |
$15.00$10.00in$30.00out |
Beaten byClaude Opus 5 | 128K | TTools2M2 modalities |
| 120Rank 120, tied with 3 others | Qwen2.5 72B InstructQwen | 1269.1 |
$0.370$0.360in$0.400outlike-for-like $0.380 |
Beaten byGemma 4 31B | 33K | TTools |
| 121Rank 121, tied with 3 others | Mistral Large 2407Mistral | 1266.3 |
$3.00$2.00in$6.00out |
Beaten byMuse Spark 1.1 | 131K | TTools2M2 modalities |
| 123Rank 123, tied with 1 others | Llama 3.1 70B InstructMeta | 1261.0 |
$0.400$0.400in$0.400outlike-for-like $0.800 |
Beaten byGemma 4 31B | 131K | TTools |
| 125Rank 125, tied with 3 others | Mistral Small 3Mistral | 1233.6 |
$0.058$0.050in$0.080out |
Beaten bySolar Pro 4 | 33K | |
| 125Rank 125, tied with 3 others | Gemma 2 27BGoogle | 1231.5 |
$0.650$0.650in$0.650out |
Beaten byMiMo-V2.5-Pro | 8K | |
| 125Rank 125, tied with 3 others | Qwen2.5 Coder 32B InstructQwen | 1230.2 |
$0.745$0.660in$1.00out |
Beaten byMiMo-V2.5-Pro | 33K | |
| 125Rank 125, tied with 3 others | Command R+ (08-2024)Cohere | 1229.0 |
$4.38$2.50in$10.00out |
Beaten byGemini 3.7 Flash | 128K | TTools |
| 129Rank 129, measurably ahead of everything below it. | Phi 4Microsoft | 1216.8 |
$0.088$0.070in$0.140out |
Beaten byQwen3 30B A3B Instruct 2507 | 16K | open |
| 130Rank 130, tied with 1 others | Claude 3 HaikuAnthropic | 1194.8 |
$0.500$0.250in$1.25out |
Beaten byGemma 4 31B | 200K | TTools2M2 modalities |
| 130Rank 130, tied with 2 others | Command R (08-2024)Cohere | 1187.5 |
$0.262$0.150in$0.600out |
Beaten byGemma 4 31B | 128K | TTools |
| 131Rank 131, tied with 1 others | Llama 3.1 8B InstructMeta | 1186.7 |
$0.058$0.050in$0.080outlike-for-like $0.220 |
Beaten bySolar Pro 4 | 131K | TTools |
| 133Rank 133, measurably ahead of everything below it. | Llama 3.2 3B InstructMeta | 1109.7 |
$0.120$0.050in$0.330out |
Beaten byDeepSeek V4 Flash 0423 | 131K | |
| 134Rank 134, measurably ahead of everything below it. | Llama 3.2 1B InstructMeta | 1054.8 |
$0.071$0.027in$0.201out |
Beaten bySolar Pro 4 | 60K |
193 models have no independent quality score. They are listed by price. Not benchmarked is not the same as benchmarked badly.
# Model General Effective $/M Beaten by Context Notes
—
unrated
$0.032$0.021in$0.063outlike-for-like $0.060
—
262K
RReasoningTTools
—
unrated
$0.055$0.030in$0.130out×3.3 past 32K+
—
1M
RReasoningTTools3M3 modalities
—
unrated
$0.075$0.060in$0.120outlike-for-like $0.079
—
1.3M
RReasoningTTools
—
unrated
$0.106$0.075in$0.200outlike-for-like $0.090
—
131K
TTools2M2 modalities
—
unrated
$0.125$0.100in$0.200out
—
1.0M
RReasoningTTools5M5 modalities
—
unrated
$0.131$0.075in$0.300out×1.3 past 128K
—
262K
RReasoningTTools3M3 modalities
—
unrated
$0.168$0.110in$0.340outlike-for-like $0.180
—
1.3M
openTTools2M2 modalities
—
unrated
$0.175$0.100in$0.400out×2 past 128K
—
262K
RReasoningTTools3M3 modalities
—
unrated
$0.175$0.100in$0.400out+ reasoning billed
—
1.0M
RReasoningTTools5M5 modalities
—
unrated
$0.202$0.117in$0.455outlike-for-like $0.250
—
262K
TTools2M2 modalities
—
unrated
$0.228$0.130in$0.520outlike-for-like $0.200
—
262K
TTools2M2 modalities
—
unrated
$0.262$0.150in$0.600outlike-for-like $0.188
—
262K
openRReasoningTTools2M2 modalities
—
unrated
$0.330$0.220in$0.660outoff-peak discount
—
1.0M
RReasoningTTools2M2 modalities
—
unrated
$0.355$0.140in$1.00outlike-for-like $0.100
—
262K
RReasoningTTools3M3 modalities
—
unrated
$0.450$0.200in$1.20out×2 past 272K×2
—
1.1M
RReasoningTTools3M3 modalities
—
unrated
$0.563$0.250in$1.50out
—
66K
RReasoning2M2 modalities
—
unrated
$0.563$0.250in$1.50out+ reasoning billed
—
1.0M
RReasoningTTools5M5 modalities
—
unrated
$0.585$0.260in$1.56out×1.3 past 256K
—
1M
RReasoningTTools3M3 modalities
—
unrated
$0.675$0.300in$1.80out×1.3 past 256K
—
1M
RReasoningTTools3M3 modalities
—
unrated
$0.688$0.250in$2.00out×2 past 128K
—
262K
RReasoningTTools3M3 modalities
—
unrated
$0.688$0.250in$2.00out×2 past 128K
—
262K
RReasoningTTools3M3 modalities
—
unrated
$0.750$0.200in$2.40outlike-for-like $0.290
—
262K
RReasoningTTools2M2 modalities
—
unrated
$0.850$0.300in$2.50out+ reasoning billed
—
33K
2M2 modalities
—
unrated
$1.13$0.500in$3.00out×2 past 128K
—
262K
RReasoningTTools3M3 modalities
—
unrated
$1.13$0.500in$3.00out
—
131K
RReasoning2M2 modalities
—
unrated
$1.13$0.500in$3.00out
—
66K
RReasoning2M2 modalities
—
unrated
$1.13$0.500in$3.00out+ reasoning billed
—
1.0M
RReasoningTTools5M5 modalities
—
unrated
$1.35$0.600in$3.60outlike-for-like $0.625
—
512K
openRReasoningTTools
—
unrated
$1.35$0.670in$3.40outlike-for-like $0.859
—
262K
openRReasoningTTools2M2 modalities
—
unrated
$1.98$1.32in$3.96outlike-for-like $1.12
—
1.0M
RReasoningTTools
—
unrated
$3.44$1.25in$10.00out×2 past 200K+ reasoning billed
—
1.0M
RReasoningTTools4M4 modalities
—
unrated
$3.44$1.25in$10.00out×2 past 200K+ reasoning billed
—
1.0M
RReasoningTTools5M5 modalities
—
unrated
$4.00$2.00in$10.00out×2 past 272K×5
—
1.1M
RReasoningTTools3M3 modalities
—
unrated
$4.50$2.00in$12.00out×2 past 272K×2
—
1.1M
RReasoningTTools3M3 modalities
—
unrated
$4.50$2.00in$12.00out+ reasoning billed
—
131K
RReasoningTTools2M2 modalities
—
unrated
$4.50$2.00in$12.00out×2 past 200K+ reasoning billed
—
1.0M
RReasoningTTools5M5 modalities
—
unrated
$4.50$2.00in$12.00out+ reasoning billed
—
66K
RReasoning2M2 modalities
—
unrated
$11.25$5.00in$30.00out×2 past 272K
—
1M
RReasoningTTools2M2 modalities
—
unrated
$20.00$10.00in$50.00out
—
1M
RReasoningTTools3M3 modalities
—
unrated
$67.50$30.00in$180out×2 past 272K×2
—
1.1M
RReasoningTTools3M3 modalities