---
title: "Best LLM for coding"
description: "Models ranked on SWE-bench, DeepSWE, LiveCodeBench, FrontierCode, FrontierSWE, LMArena and other coding evidence, with coverage shown for every row."
url: "https://aldena.ai/best-llm-for-coding"
---

# Best LLM for coding

Issue resolution, code generation, private mergeability tasks, cohort-relative engineering performance, webdev preference and security review contribute once each. LiveCodeBench is the stable 2025-08-01 snapshot of 454 problems dated 2024-08-01 through 2025-05-01, so it does not cover newer frontier families. Every row prints coverage and marks partial evidence.

Aldena runs these models inside your team rooms. [See what each one costs](https://aldena.ai/models).

| rank | model | vendor | score | coverage (benchmarks) | price (in / out) | SWE-bench Verified (%) | SWE-bench Lite (%) | SWE-bench Test (%) | SWE-bench Multimodal (%) | SWE-bench Multilingual (%) | DeepSWE v1.1 (%) | LiveCodeBench release_v6 (%) | FrontierCode 1.1 Main (%) | FrontierSWE Mean@5 Dominance (%) | MirrorCode (%) | EBR-bench (%) | LMArena coding (elo) | LMArena webdev (elo) | Warden vulnerability recall (%) | Warden total findings (count) |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | Claude Opus 5 (max) | Anthropic | 86.1 | 5 of 15 | $5.00 / $25.00 | — | — | — | — | — | 73.7% | — | 48.0% | — | — | 50.0% | 1530 | 1692 | — | — |
| 2 | Claude Opus 5 (high) | Anthropic | 82.2 | 4 of 15 | $5.00 / $25.00 | — | — | — | — | — | 72.8% | — | 48.0% | — | — | — | 1531 | 1663 | — | — |
| 3 | Claude Fable 5 | Anthropic | 81.8 | 3 of 15 | $10.00 / $50.00 | — | — | — | — | — | — | — | — | 88.2% | — | — | 1554 | 1627 | — | — |
| 4 | GPT-5.6 Sol (xhigh) | OpenAI | 80.4 | 4 of 15 | $5.00 / $30.00 | — | — | — | — | — | 70.7% | — | 46.8% | — | — | — | 1526 | 1622 | — | — |
| 5 | Kimi K3 (max) | MoonshotAI | 79.1 | 5 of 15 | $3.00 / $15.00 | — | — | — | — | — | 68.5% | — | — | — | — | — | 1543 | 1674 | 37.2% | 44.0 |
| 6 | Claude Fable 5 (high) | Anthropic | 78.8 | 3 of 15 | $10.00 / $50.00 | — | — | — | — | — | 68.6% | — | 52.7% | — | 63.9% | — | — | — | — | — |
| 7 | Claude Fable 5 (max) | Anthropic | 78.7 | 3 of 15 | $10.00 / $50.00 | — | — | — | — | — | 69.7% | — | 51.6% | — | — | 39.5% | — | — | — | — |
| 8 | Claude Opus 4.5 | Anthropic | 77.3 | 5 of 15 | $5.00 / $25.00 | 76.7% | — | 52.6% | — | 70.7% | — | — | — | — | — | — | 1523 | 1468 | — | — |
| 9 | GPT-5.6 Sol (max) | OpenAI | 77.0 | 3 of 15 | $5.00 / $30.00 | — | — | — | — | — | 72.7% | — | 47.5% | — | — | 39.0% | — | — | — | — |
| 10 | Qwen3.8 Max (partial coverage) | Qwen | 76.7 | 2 of 15 | $2.00 / $6.00 | — | — | — | — | — | — | — | — | — | — | — | 1529 | 1667 | — | — |
| 11 | Claude Fable 5 (xhigh) (partial coverage) | Anthropic | 76.3 | 2 of 15 | $10.00 / $50.00 | — | — | — | — | — | 69.9% | — | 53.5% | — | — | — | — | — | — | — |
| 12 | Claude Opus 4.6 | Anthropic | 75.8 | 5 of 15 | $5.00 / $25.00 | 78.7% | — | — | — | 72.0% | — | — | — | 48.9% | — | — | 1547 | 1537 | — | — |
| 13 | Grok 4.6 (high) | SpaceXAI | 75.5 | 4 of 15 | $2.00 / $6.00 | — | — | — | — | — | 65.2% | — | 48.0% | — | — | — | 1512 | 1631 | — | — |
| 14 | Grok 4.5 | SpaceXAI | 74.6 | 3 of 15 | $2.00 / $6.00 | — | — | — | — | — | — | — | — | 72.1% | — | — | 1521 | 1555 | — | — |
| 15 | Claude Opus 5 (medium) (partial coverage) | Anthropic | 74.3 | 2 of 15 | $5.00 / $25.00 | — | — | — | — | — | 68.9% | — | 53.4% | — | — | — | — | — | — | — |
| 16 | Muse Spark 1.1 (partial coverage) | Meta | 73.5 | 2 of 15 | $1.25 / $4.25 | — | — | — | — | — | — | — | — | — | — | — | 1531 | 1538 | — | — |
| 17 | Claude Opus 5 (xhigh) (partial coverage) | Anthropic | 73.0 | 2 of 15 | $5.00 / $25.00 | — | — | — | — | — | 73.2% | — | 43.6% | — | — | — | — | — | — | — |
| 18 | Gemini 3 Flash Preview | Google | 72.2 | 4 of 15 | $0.50 / $3.00 | 75.4% | — | — | — | 72.7% | — | — | — | — | — | — | 1508 | 1438 | — | — |
| 19 | Claude Opus 4.8 | Anthropic | 72.0 | 3 of 15 | $5.00 / $25.00 | — | — | — | — | — | — | — | — | 66.5% | — | — | 1524 | 1539 | — | — |
| 20 | Claude Opus 4.7 | Anthropic | 71.5 | 3 of 15 | $5.00 / $25.00 | — | — | — | — | — | — | — | — | 56.3% | — | — | 1547 | 1558 | — | — |
| 21 | o4 Mini High (partial coverage) | OpenAI | 71.2 | 1 of 15 | $1.10 / $4.40 | — | — | — | — | — | — | 80.2% | — | — | — | — | — | — | — | — |
| 22 | Claude Opus 4.7 (high) | Anthropic | 71.0 | 4 of 15 | $5.00 / $25.00 | — | — | — | — | — | — | — | 35.2% | — | 31.1% | — | 1552 | 1557 | — | — |
| 23 | GPT 5.5 Pre Release (xhigh) (partial coverage) | OpenAI | 70.8 | 1 of 15 | — | 80.6% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 24 | GPT-5.5 (xhigh) | OpenAI | 70.5 | 4 of 15 | $5.00 / $30.00 | — | — | — | — | — | 67.0% | — | 43.0% | — | — | 34.3% | — | 1507 | — | — |
| 25 | Claude Sonnet 4.6 | Anthropic | 70.2 | 5 of 15 | $3.00 / $15.00 | 75.2% | — | — | — | — | — | — | — | — | — | — | 1528 | 1524 | 29.1% | 32.0 |
| 26 | Claude Opus 4.5 (high 32K) (partial coverage) | Anthropic | 70.0 | 2 of 15 | $5.00 / $25.00 | — | — | — | — | — | — | — | — | — | — | — | 1530 | 1494 | — | — |
| 27 | Claude Opus 4.5 (medium) (partial coverage) | Anthropic | 70.0 | 1 of 15 | $5.00 / $25.00 | 79.2% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 28 | Qwen3.6 Max Preview | Qwen | 70.0 | 3 of 15 | $1.027 / $6.162 | 76.7% | — | — | — | — | — | — | — | — | — | — | 1509 | 1479 | — | — |
| 29 | Claude Fable 5 (medium) (partial coverage) | Anthropic | 69.9 | 2 of 15 | $10.00 / $50.00 | — | — | — | — | — | 65.4% | — | 49.8% | — | — | — | — | — | — | — |
| 30 | o3 (high) (partial coverage) | OpenAI | 69.9 | 1 of 15 | $2.00 / $8.00 | — | — | — | — | — | — | 75.8% | — | — | — | — | — | — | — | — |
| 31 | Gemini 3.7 Flash (high) | Google | 69.8 | 3 of 15 | $0.375 / $1.875 | — | — | — | — | — | 65.3% | — | 42.2% | — | — | — | — | 1587 | — | — |
| 32 | Doubao-Seed-Code (partial coverage) | ByteDance | 69.7 | 1 of 15 | — | 78.8% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 33 | GPT-5.5 (high) | OpenAI | 69.2 | 7 of 15 | $5.00 / $30.00 | — | — | — | — | — | 64.4% | — | 40.6% | — | 10.0% | — | 1520 | 1487 | 47.7% | 72.0 |
| 34 | Muse Spark (partial coverage) | Meta | 69.1 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1526 | — | — | — |
| 35 | GLM-5.3 (partial coverage) | Z.ai | 69.1 | 1 of 15 | — | — | — | — | — | — | — | — | — | 78.1% | — | — | — | — | — | — |
| 36 | Muse Spark 1.2 (xhigh) | Meta | 68.9 | 3 of 15 | $1.25 / $4.25 | — | — | — | — | — | 54.9% | — | — | — | — | — | 1533 | 1535 | — | — |
| 37 | Claude Opus 4.8 (max) (partial coverage) | Anthropic | 68.9 | 2 of 15 | $5.00 / $25.00 | — | — | — | — | — | — | — | 46.5% | — | — | 28.6% | — | — | — | — |
| 38 | DeepSeek V4 Pro (high) (partial coverage) | DeepSeek | 68.8 | 2 of 15 | $1.168 / $2.336 | — | — | — | — | — | — | — | — | — | — | — | 1489 | 1584 | — | — |
| 39 | GPT-5.6 Sol (high) | OpenAI | 68.6 | 3 of 15 | $5.00 / $30.00 | — | — | — | — | — | 69.4% | — | 45.1% | — | 20.0% | — | — | — | — | — |
| 40 | o4 Mini (medium) (partial coverage) | OpenAI | 68.5 | 1 of 15 | $1.10 / $4.40 | — | — | — | — | — | — | 74.2% | — | — | — | — | — | — | — | — |
| 41 | Hy3 (partial coverage) | Tencent | 67.8 | 2 of 15 | $0.132 / $0.528 | — | — | — | — | — | — | — | — | — | — | — | 1503 | 1522 | — | — |
| 42 | DeepSeek V4 Flash 0423 (high) (partial coverage) | DeepSeek | 67.6 | 2 of 15 | $0.0643 / $0.1285 | — | — | — | — | — | — | — | — | — | — | — | 1480 | 1581 | — | — |
| 43 | MiMo-V2.5-Pro (partial coverage) | Xiaomi | 67.6 | 2 of 15 | $0.435 / $0.87 | — | — | — | — | — | — | — | — | — | — | — | 1520 | 1474 | — | — |
| 44 | DeepSeek V4 Pro (max) (partial coverage) | DeepSeek | 67.5 | 2 of 15 | $1.168 / $2.336 | 77.6% | — | — | — | — | 62.8% | — | — | — | — | — | — | — | — | — |
| 45 | Claude Opus 4.5 (high) (partial coverage) | Anthropic | 67.4 | 1 of 15 | $5.00 / $25.00 | 76.8% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 46 | GPT-5.6 Terra (max) (partial coverage) | OpenAI | 67.2 | 2 of 15 | $1.00 / $6.00 | — | — | — | — | — | 69.6% | — | 41.3% | — | — | — | — | — | — | — |
| 47 | GPT-5.2 Chat (partial coverage) | OpenAI | 67.1 | 1 of 15 | $1.75 / $14.00 | — | — | — | — | — | — | — | — | — | — | — | 1515 | — | — | — |
| 48 | Grok 4.6 (partial coverage) | SpaceXAI | 67.0 | 1 of 15 | $2.00 / $6.00 | — | — | — | — | — | — | — | — | 77.9% | — | — | — | — | — | — |
| 49 | Qwen3.7 Max | Qwen | 66.8 | 4 of 15 | $1.475 / $4.425 | 77.3% | — | — | — | — | — | — | — | — | — | 9.5% | 1525 | 1517 | — | — |
| 50 | Ernie 5.1 (partial coverage) | Baidu | 66.7 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1514 | — | — | — |
| 51 | GPT 5.5 Instant (partial coverage) | OpenAI | 66.7 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1514 | — | — | — |
| 52 | Qwen3.5 Max Preview (partial coverage) | Qwen | 66.4 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1513 | — | — | — |
| 53 | GPT-5.4 (high) | OpenAI | 66.4 | 4 of 15 | $2.50 / $15.00 | 76.9% | — | — | — | — | — | — | — | — | 15.6% | — | 1521 | 1463 | — | — |
| 54 | Gemini 3.7 Flash (medium) (partial coverage) | Google | 66.3 | 2 of 15 | $0.375 / $1.875 | — | — | — | — | — | 65.5% | — | 43.6% | — | — | — | — | — | — | — |
| 55 | GPT-5.6 Terra (xhigh) | OpenAI | 66.2 | 4 of 15 | $1.00 / $6.00 | — | — | — | — | — | 60.2% | — | 38.8% | — | — | — | 1516 | 1521 | — | — |
| 56 | Dola Seed 2.0 Pro (partial coverage) | ByteDance | 66.1 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1513 | — | — | — |
| 57 | Claude Fable 5 (low) (partial coverage) | Anthropic | 66.1 | 2 of 15 | $10.00 / $50.00 | — | — | — | — | — | 59.6% | — | 48.0% | — | — | — | — | — | — | — |
| 58 | Gemini 3.5 Flash (medium) (partial coverage) | Google | 66.0 | 2 of 15 | $1.50 / $9.00 | — | — | — | — | — | — | — | — | — | — | — | 1507 | 1489 | — | — |
| 59 | Claude Opus 4.1 (thinking 16K) (partial coverage) | Anthropic | 66.0 | 1 of 15 | $15.00 / $75.00 | — | — | — | — | — | — | — | — | — | — | — | 1512 | — | — | — |
| 60 | Grok 4.5 (high) | SpaceXAI | 65.7 | 4 of 15 | $2.00 / $6.00 | — | — | — | — | — | 53.8% | — | 42.4% | — | — | — | — | — | 38.4% | 41.0 |
| 61 | Gemini 3 Flash Preview (high) (partial coverage) | Google | 65.7 | 1 of 15 | $0.50 / $3.00 | 75.8% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 62 | MiniMax M2.5 (high) (partial coverage) | MiniMax | 65.7 | 1 of 15 | $0.22 / $0.90 | 75.8% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 63 | Claude Opus 4.7 (max) | Anthropic | 65.5 | 3 of 15 | $5.00 / $25.00 | 83.5% | — | — | — | — | — | — | 38.5% | — | — | 19.1% | — | — | — | — |
| 64 | Grok 4.20 Multi Agent Beta 0309 (partial coverage) | xAI | 65.1 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1508 | — | — | — |
| 65 | Gemini 3.1 Pro Preview Custom Tools (partial coverage) | Google | 65.1 | 1 of 15 | $2.00 / $12.00 | 75.6% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 66 | Kimi K3 (none) (partial coverage) | MoonshotAI | 65.0 | 1 of 15 | $3.00 / $15.00 | — | — | — | — | — | — | — | 44.2% | — | — | — | — | — | — | — |
| 67 | Gemini 3 Pro | Google | 65.0 | 3 of 15 | — | — | — | — | — | 68.7% | — | — | — | — | — | — | 1518 | 1438 | — | — |
| 68 | MiniMax M3 (partial coverage) | MiniMax | 64.8 | 2 of 15 | $0.30 / $1.20 | — | — | — | — | — | — | — | — | — | — | — | 1497 | 1490 | — | — |
| 69 | Qwen3.7 Plus (partial coverage) | Qwen | 64.7 | 1 of 15 | $0.32 / $1.28 | — | — | — | — | — | — | — | — | — | — | — | 1506 | — | — | — |
| 70 | EXAONE 4.0 32B (partial coverage) | LG AI Research | 64.5 | 1 of 15 | — | — | — | — | — | — | — | 70.0% | — | — | — | — | — | — | — | — |
| 71 | Grok 4.6 (medium) (partial coverage) | SpaceXAI | 64.4 | 1 of 15 | $2.00 / $6.00 | — | — | — | — | — | 67.5% | — | — | — | — | — | — | — | — | — |
| 72 | GPT-5.5 | OpenAI | 64.3 | 3 of 15 | $5.00 / $30.00 | — | — | — | — | — | — | — | — | 64.5% | — | — | 1510 | 1457 | — | — |
| 73 | GLM 5 | Z.ai | 64.1 | 4 of 15 | $0.60 / $1.92 | 72.1% | — | — | — | 69.7% | — | — | — | — | — | — | 1497 | 1436 | — | — |
| 74 | GPT-5.3-Codex (high) (partial coverage) | OpenAI | 64.0 | 1 of 15 | $1.75 / $14.00 | 74.8% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 75 | Seed 2.1 Pro Preview (partial coverage) | ByteDance | 64.0 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | — | 1522 | — | — |
| 76 | R1 0528 (partial coverage) | DeepSeek | 63.7 | 2 of 15 | $0.50 / $2.15 | — | — | — | — | — | — | 73.1% | — | — | — | — | 1464 | — | — | — |
| 77 | GPT-5 (partial coverage) | OpenAI | 63.6 | 1 of 15 | $1.25 / $10.00 | 74.4% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 78 | Claude Sonnet 5 (high) | Anthropic | 63.5 | 4 of 15 | $2.00 / $10.00 | — | — | — | — | — | 48.2% | — | 39.4% | — | — | — | 1523 | 1540 | — | — |
| 79 | Claude Opus 4 (thinking 16K) (partial coverage) | Anthropic | 63.5 | 1 of 15 | $15.00 / $75.00 | — | — | — | — | — | — | — | — | — | — | — | 1499 | — | — | — |
| 80 | Gemini 3.5 Flash Lite (partial coverage) | Google | 63.4 | 2 of 15 | $0.30 / $2.50 | — | — | — | — | — | — | — | — | — | — | — | 1503 | 1449 | — | — |
| 81 | OpenReasoning Nemotron 32B (partial coverage) | NVIDIA | 63.2 | 1 of 15 | — | — | — | — | — | — | — | 69.8% | — | — | — | — | — | — | — | — |
| 82 | GPT-5.6 Luna (max) (partial coverage) | OpenAI | 63.0 | 2 of 15 | $0.10 / $0.60 | — | — | — | — | — | 67.2% | — | 39.8% | — | — | — | — | — | — | — |
| 83 | GLM 5.1 | Z.ai | 62.9 | 4 of 15 | $0.966 / $3.036 | 74.2% | — | — | — | — | — | — | — | 25.7% | — | — | 1515 | 1510 | — | — |
| 84 | GLM 5.2 (partial coverage) | Z.ai | 62.9 | 1 of 15 | $0.462 / $1.452 | — | — | — | — | — | — | — | — | 67.5% | — | — | — | — | — | — |
| 85 | GPT-5.6 Luna (xhigh) | OpenAI | 62.8 | 4 of 15 | $0.10 / $0.60 | — | — | — | — | — | 56.9% | — | 38.9% | — | — | — | 1499 | 1517 | — | — |
| 86 | Grok 4.6 (xhigh) (partial coverage) | SpaceXAI | 62.7 | 1 of 15 | $2.00 / $6.00 | — | — | — | — | — | 66.7% | — | — | — | — | — | — | — | — | — |
| 87 | GPT-5.3 Chat (partial coverage) | OpenAI | 62.5 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1496 | — | — | — |
| 88 | Claude Opus 4.8 (xhigh) (partial coverage) | Anthropic | 62.2 | 2 of 15 | $5.00 / $25.00 | — | — | — | — | — | 54.4% | — | 45.5% | — | — | — | — | — | — | — |
| 89 | Grok 4.1 (partial coverage) | xAI | 62.1 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1492 | — | — | — |
| 90 | GPT-5.6 Luna (high) | OpenAI | 62.1 | 4 of 15 | $0.10 / $0.60 | — | — | — | — | — | 44.3% | — | 35.9% | — | — | — | — | — | 41.9% | 57.0 |
| 91 | SWE-1.7 (none) (partial coverage) | Cognition | 61.5 | 1 of 15 | — | — | — | — | — | — | — | — | 42.0% | — | — | — | — | — | — | — |
| 92 | Claude Sonnet 4.5 (high 32K) (partial coverage) | Anthropic | 61.5 | 2 of 15 | $3.00 / $15.00 | — | — | — | — | — | — | — | — | — | — | — | 1519 | 1392 | — | — |
| 93 | Kimi K2.5 (thinking) (partial coverage) | MoonshotAI | 61.5 | 2 of 15 | $0.57 / $2.85 | — | — | — | — | — | — | — | — | — | — | — | 1502 | 1436 | — | — |
| 94 | o3 | OpenAI | 61.5 | 3 of 15 | $2.00 / $8.00 | 58.4% | — | — | 36.0% | — | — | — | — | — | — | — | 1460 | — | — | — |
| 95 | GLM 5.2 (max) | Z.ai | 61.4 | 5 of 15 | $0.462 / $1.452 | 78.7% | — | — | — | — | 43.8% | — | — | — | — | 9.5% | 1506 | 1585 | — | — |
| 96 | Gemini 3 Pro Preview (partial coverage) | Google | 61.3 | 1 of 15 | — | 72.9% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 97 | Mimo v2 Pro (partial coverage) | Xiaomi | 61.2 | 2 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1503 | 1434 | — | — |
| 98 | Ernie 5.0 0110 (partial coverage) | Baidu | 61.2 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1490 | — | — | — |
| 99 | GLM 5 (high) (partial coverage) | Z.ai | 60.8 | 1 of 15 | $0.60 / $1.92 | 72.8% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 100 | Mimo v2 Omni (partial coverage) | Xiaomi | 60.8 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1486 | — | — | — |
| 101 | Grok 4.5 (medium) (partial coverage) | SpaceXAI | 60.7 | 1 of 15 | $2.00 / $6.00 | — | — | — | — | — | — | — | 41.9% | — | — | — | — | — | — | — |
| 102 | MiMo-V2.5 (partial coverage) | Xiaomi | 60.6 | 2 of 15 | $0.14 / $0.28 | — | — | — | — | — | — | — | — | — | — | — | 1491 | 1438 | — | — |
| 103 | OpenCodeReasoning Nemotron 1.1 32B (partial coverage) | NVIDIA | 60.5 | 1 of 15 | — | — | — | — | — | — | — | 66.8% | — | — | — | — | — | — | — | — |
| 104 | Kimi K2.5 Instant (partial coverage) | Moonshot AI | 60.4 | 2 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1505 | 1405 | — | — |
| 105 | DeepSeek V4 Flash 0423 (partial coverage) | DeepSeek | 60.4 | 1 of 15 | $0.0643 / $0.1285 | — | — | — | — | — | — | — | — | — | — | — | 1483 | — | — | — |
| 106 | Claude 3.7 Sonnet | Anthropic | 60.3 | 5 of 15 | — | 61.0% | 51.7% | 33.8% | 31.3% | — | — | — | — | — | — | — | 1430 | — | — | — |
| 107 | Claude Opus 5 (low) (partial coverage) | Anthropic | 60.2 | 2 of 15 | $5.00 / $25.00 | — | — | — | — | — | 58.1% | — | 42.0% | — | — | — | — | — | — | — |
| 108 | GPT-5.1 (high) (partial coverage) | OpenAI | 59.9 | 2 of 15 | $1.25 / $10.00 | 68.0% | — | — | — | — | — | — | — | — | — | — | 1491 | — | — | — |
| 109 | GPT-5.6 Sol (medium) (partial coverage) | OpenAI | 59.5 | 2 of 15 | $5.00 / $30.00 | — | — | — | — | — | 61.1% | — | 39.9% | — | — | — | — | — | — | — |
| 110 | Claude Sonnet 4.5 (high) (partial coverage) | Anthropic | 59.5 | 1 of 15 | $3.00 / $15.00 | 71.4% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 111 | GPT-5.2 (xhigh) (partial coverage) | OpenAI | 59.4 | 1 of 15 | $1.75 / $14.00 | — | — | — | — | — | — | — | — | — | — | 23.0% | — | — | — | — |
| 112 | Nemotron 3 Ultra (partial coverage) | NVIDIA | 59.3 | 1 of 15 | $0.60 / $3.60 | — | — | — | — | — | — | — | — | — | — | — | 1475 | — | — | — |
| 113 | Gemini 3.6 Flash (high) | Google | 59.1 | 4 of 15 | $0.75 / $3.75 | — | — | — | — | — | 46.7% | — | 33.7% | — | — | — | 1522 | 1537 | — | — |
| 114 | DeepSeek V3.2 Exp (thinking) (partial coverage) | DeepSeek | 59.0 | 1 of 15 | $0.27 / $0.41 | — | — | — | — | — | — | — | — | — | — | — | 1475 | — | — | — |
| 115 | Gemini 3.1 Pro Preview | Google | 58.8 | 4 of 15 | $2.00 / $12.00 | — | — | — | — | — | — | — | — | 34.4% | — | 14.3% | 1521 | 1447 | — | — |
| 116 | Claude Sonnet 4 | Anthropic | 58.8 | 5 of 15 | $3.00 / $15.00 | 57.0% | 58.3% | — | 35.6% | — | — | 47.1% | — | — | — | — | 1449 | — | — | — |
| 117 | Kimi K2 0905 (partial coverage) | MoonshotAI | 58.7 | 2 of 15 | $0.60 / $2.50 | 71.2% | — | — | — | — | — | — | — | — | — | — | 1468 | — | — | — |
| 118 | LongCat Flash Chat (partial coverage) | Meituan | 58.7 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1474 | — | — | — |
| 119 | Claude Sonnet 5 (max) (partial coverage) | Anthropic | 58.7 | 2 of 15 | $2.00 / $10.00 | — | — | — | — | — | 53.9% | — | 42.4% | — | — | — | — | — | — | — |
| 120 | GLM 4.7 (partial coverage) | Z.ai | 58.7 | 2 of 15 | $0.40 / $1.75 | — | — | — | — | — | — | — | — | — | — | — | 1485 | 1434 | — | — |
| 121 | Claude Sonnet 4 (thinking 32K) (partial coverage) | Anthropic | 58.6 | 1 of 15 | $3.00 / $15.00 | — | — | — | — | — | — | — | — | — | — | — | 1473 | — | — | — |
| 122 | Qwen3 Max (partial coverage) | Qwen | 58.4 | 1 of 15 | $0.78 / $3.90 | — | — | — | — | — | — | — | — | — | — | — | 1473 | — | — | — |
| 123 | GPT-5.2 (high) | OpenAI | 58.3 | 3 of 15 | $1.75 / $14.00 | 73.8% | — | — | — | 66.7% | — | — | — | — | — | — | 1490 | — | — | — |
| 124 | Kimi K2.5 (high) (partial coverage) | MoonshotAI | 58.3 | 1 of 15 | $0.57 / $2.85 | 70.8% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 125 | Qwen3 235B A22B Instruct 2507 (partial coverage) | Qwen | 58.3 | 1 of 15 | $0.09 / $0.55 | — | — | — | — | — | — | — | — | — | — | — | 1472 | — | — | — |
| 126 | Grok 4.20 Beta 0309 (reasoning) (partial coverage) | xAI | 58.3 | 2 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1511 | 1374 | — | — |
| 127 | Qwen3.5 397B A17B (partial coverage) | Qwen | 58.2 | 2 of 15 | $0.39 / $2.34 | — | — | — | — | — | — | — | — | — | — | — | 1491 | 1400 | — | — |
| 128 | Ernie 5.0 Preview 1203 (partial coverage) | Baidu | 58.2 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1472 | — | — | — |
| 129 | Claude Opus 4.6 (high) | Anthropic | 57.9 | 5 of 15 | $5.00 / $25.00 | — | — | — | — | — | — | — | 26.6% | — | — | — | 1552 | 1545 | 26.7% | 24.0 |
| 130 | Chatgpt 4o (partial coverage) | OpenAI | 57.8 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1468 | — | — | — |
| 131 | GPT-5 (medium) (partial coverage) | OpenAI | 57.7 | 2 of 15 | $1.25 / $10.00 | 71.5% | — | — | — | — | — | — | — | — | — | — | — | 1419 | — | — |
| 132 | Claude Opus 4.8 (high) | Anthropic | 57.6 | 6 of 15 | $5.00 / $25.00 | — | — | — | — | — | 51.8% | — | 41.0% | — | — | — | 1533 | 1564 | 24.4% | 24.0 |
| 133 | GLM 5V Turbo (partial coverage) | Z.ai | 57.6 | 2 of 15 | $1.20 / $4.00 | — | — | — | — | — | — | — | — | — | — | — | 1490 | 1400 | — | — |
| 134 | DeepSeek V3.2 (high) (partial coverage) | DeepSeek | 57.6 | 1 of 15 | $0.269 / $0.40 | 70.0% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 135 | GPT-5.4 (medium) (partial coverage) | OpenAI | 57.4 | 1 of 15 | $2.50 / $15.00 | — | — | — | — | — | — | — | — | — | — | — | — | 1442 | — | — |
| 136 | GPT-5.2 | OpenAI | 57.4 | 3 of 15 | $1.75 / $14.00 | 69.0% | — | — | — | — | — | — | — | — | — | — | 1482 | 1418 | — | — |
| 137 | o4 Mini (low) (partial coverage) | OpenAI | 57.2 | 1 of 15 | $1.10 / $4.40 | — | — | — | — | — | — | 65.9% | — | — | — | — | — | — | — | — |
| 138 | Qwen3 VL 235B A22B Instruct (partial coverage) | Qwen | 57.1 | 1 of 15 | $0.26 / $1.04 | — | — | — | — | — | — | — | — | — | — | — | 1465 | — | — | — |
| 139 | Gemini 3 Pro Preview (high) (partial coverage) | Google | 57.0 | 1 of 15 | — | 69.6% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 140 | GPT-5.4 | OpenAI | 57.0 | 3 of 15 | $2.50 / $15.00 | — | — | — | — | — | — | — | — | 46.0% | — | — | 1514 | 1390 | — | — |
| 141 | GPT-5 (high) | OpenAI | 56.9 | 3 of 15 | $1.25 / $10.00 | 73.5% | — | — | — | — | — | — | — | — | — | 12.7% | 1469 | — | — | — |
| 142 | DeepSeek V3.1 Terminus (thinking) (partial coverage) | DeepSeek | 56.6 | 1 of 15 | $0.27 / $0.95 | — | — | — | — | — | — | — | — | — | — | — | 1463 | — | — | — |
| 143 | GPT 5 Chat (partial coverage) | OpenAI | 56.5 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1462 | — | — | — |
| 144 | Qwen3.8 Max (xhigh) (partial coverage) | Qwen | 56.5 | 1 of 15 | $2.00 / $6.00 | — | — | — | — | — | 57.5% | — | — | — | — | — | — | — | — | — |
| 145 | o3 Mini High (partial coverage) | OpenAI | 56.4 | 2 of 15 | $1.10 / $4.40 | — | — | — | — | — | — | 67.4% | — | — | — | — | 1435 | — | — | — |
| 146 | Gemini 3.5 Flash (high) | Google | 56.1 | 5 of 15 | $1.50 / $9.00 | 79.3% | — | — | — | — | 36.1% | — | — | — | — | 4.8% | 1509 | 1499 | — | — |
| 147 | MiniMax M2.7 (partial coverage) | MiniMax | 56.0 | 2 of 15 | $0.30 / $1.20 | — | — | — | — | — | — | — | — | — | — | — | 1479 | 1397 | — | — |
| 148 | Gemma 4 31B (partial coverage) | Google | 56.0 | 2 of 15 | $0.10 / $0.34 | — | — | — | — | — | — | — | — | — | — | — | 1499 | 1364 | — | — |
| 149 | GPT-5.4 Nano (high) (partial coverage) | OpenAI | 56.0 | 1 of 15 | $0.20 / $1.25 | — | — | — | — | — | — | — | — | — | — | — | 1460 | — | — | — |
| 150 | Claude Sonnet 5 (xhigh) (partial coverage) | Anthropic | 56.0 | 2 of 15 | $2.00 / $10.00 | — | — | — | — | — | 49.7% | — | 42.7% | — | — | — | — | — | — | — |
| 151 | Grok 4.5 (low) (partial coverage) | SpaceXAI | 55.8 | 1 of 15 | $2.00 / $6.00 | — | — | — | — | — | — | — | 37.9% | — | — | — | — | — | — | — |
| 152 | Claude Haiku 4.5 (high) (partial coverage) | Anthropic | 55.7 | 1 of 15 | $1.00 / $5.00 | 66.6% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 153 | GPT 4.5 Preview (partial coverage) | OpenAI | 55.5 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1459 | — | — | — |
| 154 | Gemini 3 Flash Preview (thinking minimal) (partial coverage) | Google | 55.4 | 2 of 15 | $0.50 / $3.00 | — | — | — | — | — | — | — | — | — | — | — | 1491 | 1383 | — | — |
| 155 | XBai o4 (medium) (partial coverage) | MetaStoneTec | 55.2 | 1 of 15 | — | — | — | — | — | — | — | 65.0% | — | — | — | — | — | — | — | — |
| 156 | GPT-5.4 (xhigh) (partial coverage) | OpenAI | 55.1 | 2 of 15 | $2.50 / $15.00 | — | — | — | — | — | 51.8% | — | — | — | — | 25.4% | — | — | — | — |
| 157 | GPT-5.1-Codex (medium) (partial coverage) | OpenAI | 55.1 | 1 of 15 | $1.25 / $10.00 | 66.0% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 158 | DeepSeek V3.1 (thinking) (partial coverage) | DeepSeek | 55.0 | 1 of 15 | $0.25 / $0.95 | — | — | — | — | — | — | — | — | — | — | — | 1457 | — | — | — |
| 159 | Mistral Medium 2508 (partial coverage) | Mistral AI | 54.7 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1455 | — | — | — |
| 160 | Qwen3 VL 235B A22B Thinking (partial coverage) | Qwen | 54.6 | 1 of 15 | $0.40 / $4.00 | — | — | — | — | — | — | — | — | — | — | — | 1455 | — | — | — |
| 161 | Kimi K2 0711 (partial coverage) | MoonshotAI | 54.6 | 2 of 15 | $0.57 / $2.30 | 65.4% | — | — | — | — | — | — | — | — | — | — | 1461 | — | — | — |
| 162 | Claude Opus 4.5 (128K) (partial coverage) | Anthropic | 54.5 | 1 of 15 | $5.00 / $25.00 | — | — | — | — | — | — | — | — | — | — | 14.3% | — | — | — | — |
| 163 | Claude Sonnet 4.5 | Anthropic | 54.5 | 6 of 15 | $3.00 / $15.00 | 71.3% | — | 44.3% | — | 67.0% | — | — | — | — | — | 2.4% | 1513 | 1386 | — | — |
| 164 | GPT-5.3-Codex (partial coverage) | OpenAI | 54.4 | 1 of 15 | $1.75 / $14.00 | — | — | — | — | — | — | — | — | — | — | — | — | 1409 | — | — |
| 165 | DeepSeek V4 Pro (xhigh) (partial coverage) | DeepSeek | 54.4 | 2 of 15 | $1.168 / $2.336 | — | — | — | — | — | — | — | — | — | — | — | — | — | 26.7% | 30.0 |
| 166 | Claude 3.7 Sonnet (thinking 32K) (partial coverage) | Anthropic | 54.3 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1452 | — | — | — |
| 167 | Step 3.5 Flash (partial coverage) | StepFun | 54.2 | 1 of 15 | $0.10 / $0.30 | — | — | — | — | — | — | — | — | — | — | — | 1451 | — | — | — |
| 168 | GPT-5 Mini (medium) (partial coverage) | OpenAI | 54.1 | 1 of 15 | $0.25 / $2.00 | 64.7% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 169 | DeepSeek V4 Pro | DeepSeek | 53.9 | 3 of 15 | $1.168 / $2.336 | — | — | — | — | — | — | — | — | 24.6% | — | — | 1502 | 1446 | — | — |
| 170 | Claude Opus 4.1 | Anthropic | 53.9 | 4 of 15 | $15.00 / $75.00 | 73.3% | — | — | — | — | — | — | — | — | — | 7.9% | 1505 | 1389 | — | — |
| 171 | DeepSeek V3.1 (partial coverage) | DeepSeek | 53.6 | 1 of 15 | $0.25 / $0.95 | — | — | — | — | — | — | — | — | — | — | — | 1448 | — | — | — |
| 172 | Claude 3.5 Haiku (partial coverage) | Anthropic | 53.5 | 2 of 15 | — | — | 41.7% | — | — | — | — | — | — | — | — | — | 1385 | — | — | — |
| 173 | Kimi K2 Thinking (partial coverage) | MoonshotAI | 53.4 | 1 of 15 | $0.60 / $2.50 | 63.4% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 174 | Qwen3 Next 80B A3B Instruct (partial coverage) | Qwen | 53.4 | 1 of 15 | $0.10 / $1.10 | — | — | — | — | — | — | — | — | — | — | — | 1446 | — | — | — |
| 175 | Gemma 4 26B A4B  (partial coverage) | Google | 53.3 | 2 of 15 | $0.12 / $0.40 | — | — | — | — | — | — | — | — | — | — | — | 1481 | 1362 | — | — |
| 176 | Qwen3 235B A22B (nothinking) (partial coverage) | Qwen | 53.2 | 1 of 15 | $0.455 / $1.82 | — | — | — | — | — | — | — | — | — | — | — | 1446 | — | — | — |
| 177 | GPT-5.5 (medium) (partial coverage) | OpenAI | 53.2 | 2 of 15 | $5.00 / $30.00 | — | — | — | — | — | 54.0% | — | 36.6% | — | — | — | — | — | — | — |
| 178 | R1 (partial coverage) | DeepSeek | 53.1 | 1 of 15 | $0.70 / $2.50 | — | — | — | — | — | — | — | — | — | — | — | 1445 | — | — | — |
| 179 | Muse Glimmer 30B (partial coverage) | Meta | 52.9 | 2 of 15 | $0.35 / $1.50 | — | — | — | — | — | — | — | — | — | — | — | 1481 | 1359 | — | — |
| 180 | Kimi K2.6 | MoonshotAI | 52.9 | 5 of 15 | $0.5415 / $2.28 | 76.7% | — | — | — | — | — | — | — | 22.2% | — | 2.4% | 1514 | 1509 | — | — |
| 181 | Grok 3 Beta (partial coverage) | xAI | 52.8 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1443 | — | — | — |
| 182 | Trinity Large Preview (partial coverage) | Arcee AI | 52.7 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1443 | — | — | — |
| 183 | Grok 4.3 (partial coverage) | SpaceXAI | 52.7 | 2 of 15 | $1.25 / $2.50 | — | — | — | — | — | — | — | — | — | — | — | 1488 | 1354 | — | — |
| 184 | o3 (medium) (partial coverage) | OpenAI | 52.6 | 1 of 15 | $2.00 / $8.00 | 62.3% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 185 | Gemini 3.7 Flash (low) (partial coverage) | Google | 52.6 | 2 of 15 | $0.375 / $1.875 | — | — | — | — | — | 53.8% | — | 36.9% | — | — | — | — | — | — | — |
| 186 | Qwen3 235B A22B Thinking 2507 (partial coverage) | Qwen | 52.5 | 1 of 15 | $0.23 / $2.30 | — | — | — | — | — | — | — | — | — | — | — | 1442 | — | — | — |
| 187 | Qwen3 235B A22B (partial coverage) | Qwen | 52.5 | 2 of 15 | $0.455 / $1.82 | — | — | — | — | — | — | 65.9% | — | — | — | — | 1433 | — | — | — |
| 188 | Qwen3 30B A3B Instruct 2507 (partial coverage) | Qwen | 52.3 | 1 of 15 | $0.0482 / $0.1931 | — | — | — | — | — | — | — | — | — | — | — | 1440 | — | — | — |
| 189 | GPT-5.6 Terra (high) (partial coverage) | OpenAI | 52.2 | 2 of 15 | $1.00 / $6.00 | — | — | — | — | — | 53.8% | — | 36.9% | — | — | — | — | — | — | — |
| 190 | DeepSeek V3.1 Terminus (partial coverage) | DeepSeek | 52.1 | 1 of 15 | $0.27 / $0.95 | — | — | — | — | — | — | — | — | — | — | — | 1439 | — | — | — |
| 191 | Hunyuan Vision 1.5 (thinking) (partial coverage) | Tencent | 52.0 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1438 | — | — | — |
| 192 | GPT-5.1 (medium) (partial coverage) | OpenAI | 51.9 | 2 of 15 | $1.25 / $10.00 | 66.0% | — | — | — | — | — | — | — | — | — | — | — | 1391 | — | — |
| 193 | Claude Opus 4.7 (xhigh) (partial coverage) | Anthropic | 51.9 | 1 of 15 | $5.00 / $25.00 | — | — | — | — | — | — | — | 34.9% | — | — | — | — | — | — | — |
| 194 | Gemini 3.6 Flash (medium) (partial coverage) | Google | 51.4 | 1 of 15 | $0.75 / $3.75 | — | — | — | — | — | — | — | 34.4% | — | — | — | — | — | — | — |
| 195 | Grok 4 0709 (partial coverage) | xAI | 51.3 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1435 | — | — | — |
| 196 | o3 Mini (low) (partial coverage) | OpenAI | 51.2 | 1 of 15 | $1.10 / $4.40 | — | — | — | — | — | — | 57.0% | — | — | — | — | — | — | — | — |
| 197 | DeepSeek V4 Flash 0423 (max) (partial coverage) | DeepSeek | 51.1 | 1 of 15 | $0.0643 / $0.1285 | — | — | — | — | — | 53.3% | — | — | — | — | — | — | — | — | — |
| 198 | Muse Spark 1.1 (xhigh) (partial coverage) | Meta | 51.1 | 1 of 15 | $1.25 / $4.25 | — | — | — | — | — | 53.3% | — | — | — | — | — | — | — | — | — |
| 199 | MiniMax M2.5 | MiniMax | 51.0 | 3 of 15 | $0.22 / $0.90 | — | — | — | — | 68.3% | — | — | — | — | — | — | 1444 | 1384 | — | — |
| 200 | Claude Opus 4 | Anthropic | 50.9 | 3 of 15 | $15.00 / $75.00 | 70.7% | — | — | — | — | — | 46.9% | — | — | — | — | 1464 | — | — | — |
| 201 | Mistral Medium 2505 (partial coverage) | Mistral AI | 50.9 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1433 | — | — | — |
| 202 | GPT-5.1 (partial coverage) | OpenAI | 50.7 | 2 of 15 | $1.25 / $10.00 | — | — | — | — | — | — | — | — | — | — | — | 1474 | 1341 | — | — |
| 203 | Gemini 2.5 Pro | Google | 50.6 | 4 of 15 | $1.25 / $10.00 | 57.6% | — | — | — | — | — | 73.6% | — | — | — | — | 1465 | 1226 | — | — |
| 204 | Claude Opus 4.6 (max) (partial coverage) | Anthropic | 50.6 | 1 of 15 | $5.00 / $25.00 | — | — | — | — | — | — | — | — | — | — | 12.7% | — | — | — | — |
| 205 | Grok 3 Mini Beta (high) (partial coverage) | xAI | 50.6 | 2 of 15 | — | — | — | — | — | — | — | 66.7% | — | — | — | — | 1390 | — | — | — |
| 206 | Kimi K2 Thinking Turbo (partial coverage) | Moonshot AI | 50.4 | 2 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1486 | 1323 | — | — |
| 207 | Ernie 5.0 Preview 1022 (partial coverage) | Baidu | 50.3 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1432 | — | — | — |
| 208 | DeepSeek V3.2 (thinking) | DeepSeek | 50.1 | 3 of 15 | $0.269 / $0.40 | 60.0% | — | — | — | — | — | — | — | — | — | — | 1475 | 1360 | — | — |
| 209 | GPT-5.4 Mini (high) | OpenAI | 50.1 | 3 of 15 | $0.75 / $4.50 | — | — | — | — | — | — | — | 23.1% | — | — | — | 1497 | 1397 | — | — |
| 210 | Kimi K2.5 | MoonshotAI | 50.1 | 3 of 15 | $0.57 / $2.85 | 73.8% | — | — | — | 67.3% | — | — | — | 23.4% | — | — | — | — | — | — |
| 211 | GPT-5 Mini (high) (partial coverage) | OpenAI | 50.1 | 1 of 15 | $0.25 / $2.00 | — | — | — | — | — | — | — | — | — | — | — | 1431 | — | — | — |
| 212 | o1 (partial coverage) | OpenAI | 50.0 | 2 of 15 | $15.00 / $60.00 | 64.6% | — | — | — | — | — | — | — | — | — | — | 1433 | — | — | — |
| 213 | Claude Opus 4 (thinking) (partial coverage) | Anthropic | 49.9 | 1 of 15 | $15.00 / $75.00 | — | — | — | — | — | — | 56.6% | — | — | — | — | — | — | — | — |
| 214 | o4 Mini | OpenAI | 49.8 | 3 of 15 | $1.10 / $4.40 | 45.0% | — | — | 33.9% | — | — | — | — | — | — | — | 1433 | — | — | — |
| 215 | DeepSeek V3 0324 (partial coverage) | DeepSeek | 49.8 | 1 of 15 | $0.27 / $1.12 | — | — | — | — | — | — | — | — | — | — | — | 1429 | — | — | — |
| 216 | Kimi K2.7 Code (none) (partial coverage) | MoonshotAI | 49.7 | 1 of 15 | $0.71 / $3.50 | — | — | — | — | — | — | — | 30.1% | — | — | — | — | — | — | — |
| 217 | GLM 4.5 Air (partial coverage) | Z.ai | 49.6 | 1 of 15 | $0.13 / $0.85 | — | — | — | — | — | — | — | — | — | — | — | 1426 | — | — | — |
| 218 | Devstral Small 2512 (partial coverage) | Mistral AI | 49.6 | 1 of 15 | — | 56.4% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 219 | GLM 4.7 Flash (partial coverage) | Z.ai | 49.5 | 1 of 15 | $0.06 / $0.40 | — | — | — | — | — | — | — | — | — | — | — | 1424 | — | — | — |
| 220 | Qwen3.5-122B-A10B (partial coverage) | Qwen | 49.4 | 2 of 15 | $0.29 / $2.40 | — | — | — | — | — | — | — | — | — | — | — | 1459 | 1358 | — | — |
| 221 | Hunyuan Hy3 Preview (partial coverage) | Tencent | 49.4 | 2 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1461 | 1356 | — | — |
| 222 | Solar Pro 4 (partial coverage) | Upstage | 49.3 | 2 of 15 | $0.03 / $0.12 | — | — | — | — | — | — | — | — | — | — | — | 1450 | 1371 | — | — |
| 223 | GPT-5.5 (low) | OpenAI | 49.3 | 4 of 15 | $5.00 / $30.00 | — | — | — | — | — | 27.0% | — | 30.6% | — | — | — | — | — | 32.6% | 38.0 |
| 224 | Nemotron 3.5 Lightning 30B A3B (partial coverage) | NVIDIA | 49.2 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1422 | — | — | — |
| 225 | MiniMax M2.1 (partial coverage) | MiniMax | 49.2 | 2 of 15 | $0.30 / $1.20 | — | — | — | — | — | — | — | — | — | — | — | 1440 | 1387 | — | — |
| 226 | Qwen3 Next 80B A3B Thinking (partial coverage) | Qwen | 49.1 | 1 of 15 | $0.15 / $1.20 | — | — | — | — | — | — | — | — | — | — | — | 1421 | — | — | — |
| 227 | GLM 4.6V (partial coverage) | Z.ai | 49.0 | 1 of 15 | $0.30 / $0.90 | — | — | — | — | — | — | — | — | — | — | — | 1417 | — | — | — |
| 228 | Grok 4.1 (thinking) (partial coverage) | xAI | 48.9 | 2 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1499 | 1210 | — | — |
| 229 | GLM 4.5 (partial coverage) | Z.ai | 48.8 | 2 of 15 | $0.60 / $2.20 | 54.2% | — | — | — | — | — | — | — | — | — | — | 1455 | — | — | — |
| 230 | Claude 3.5 Sonnet | Anthropic | 48.8 | 6 of 15 | — | 62.8% | 51.3% | 24.9% | 25.3% | — | — | 36.4% | — | — | — | — | 1435 | — | — | — |
| 231 | MiniMax M1 (partial coverage) | MiniMax | 48.7 | 1 of 15 | $0.55 / $2.20 | — | — | — | — | — | — | — | — | — | — | — | 1416 | — | — | — |
| 232 | Claude Sonnet 5 (partial coverage) | Anthropic | 48.6 | 2 of 15 | $2.00 / $10.00 | — | — | — | — | — | — | — | — | — | — | — | — | — | 25.6% | 27.0 |
| 233 | Claude Sonnet 4 (thinking) (partial coverage) | Anthropic | 48.5 | 1 of 15 | $3.00 / $15.00 | — | — | — | — | — | — | 56.0% | — | — | — | — | — | — | — | — |
| 234 | Mistral Small 2506 (partial coverage) | Mistral AI | 48.4 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1412 | — | — | — |
| 235 | Claude Opus 4.7 (low) (partial coverage) | Anthropic | 48.4 | 1 of 15 | $5.00 / $25.00 | — | — | — | — | — | — | — | 27.6% | — | — | — | — | — | — | — |
| 236 | Ling Flash 2.0 (partial coverage) | inclusionAI | 48.3 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1411 | — | — | — |
| 237 | Composer 2.5 (partial coverage) | Cursor | 48.3 | 1 of 15 | — | — | — | — | — | — | — | — | — | 33.8% | — | — | — | — | — | — |
| 238 | Qwen3.6 Plus | Qwen | 48.1 | 4 of 15 | $0.325 / $1.95 | 57.9% | — | — | — | — | — | — | — | 19.9% | — | — | 1495 | 1460 | — | — |
| 239 | Mistral Medium 3.5 (partial coverage) | Mistral | 48.1 | 2 of 15 | $1.50 / $7.50 | — | — | — | — | — | — | — | — | — | — | — | 1479 | 1265 | — | — |
| 240 | INTELLECT-3 (partial coverage) | Prime Intellect | 48.1 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1409 | — | — | — |
| 241 | Step 3 (partial coverage) | StepFun | 48.0 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1408 | — | — | — |
| 242 | Inkling | Thinking Machines | 48.0 | 3 of 15 | $0.95 / $4.05 | — | — | — | — | — | — | — | 14.0% | — | — | — | 1494 | 1405 | — | — |
| 243 | GPT-5.4 Mini (xhigh) (partial coverage) | OpenAI | 47.9 | 1 of 15 | $0.75 / $4.50 | — | — | — | — | — | — | — | 27.0% | — | — | — | — | — | — | — |
| 244 | Qwen3 Coder 480B A35b Instruct | Qwen | 47.9 | 3 of 15 | — | 69.6% | — | — | — | — | — | — | — | — | — | — | 1457 | 1273 | — | — |
| 245 | Nemotron 3 Super (partial coverage) | NVIDIA | 47.9 | 1 of 15 | $0.085 / $0.40 | — | — | — | — | — | — | — | — | — | — | — | 1408 | — | — | — |
| 246 | Qwen3.5-27B (partial coverage) | Qwen | 47.9 | 2 of 15 | $0.195 / $1.56 | — | — | — | — | — | — | — | — | — | — | — | 1450 | 1358 | — | — |
| 247 | Kimi K2.7 Code (partial coverage) | MoonshotAI | 47.8 | 2 of 15 | $0.71 / $3.50 | — | — | — | — | — | 30.5% | — | — | — | — | — | — | 1473 | — | — |
| 248 | Qwen3 32B (partial coverage) | Qwen | 47.7 | 1 of 15 | $0.08 / $0.28 | — | — | — | — | — | — | — | — | — | — | — | 1407 | — | — | — |
| 249 | Qwen3 Coder 30B A3B Instruct (partial coverage) | Qwen | 47.7 | 1 of 15 | $0.07 / $0.28 | 51.6% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 250 | GLM 4.5V (partial coverage) | Z.ai | 47.6 | 1 of 15 | $0.60 / $1.80 | — | — | — | — | — | — | — | — | — | — | — | 1405 | — | — | — |
| 251 | Llama 3.3 Nemotron Super 49B v1.5 (partial coverage) | NVIDIA | 47.5 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1404 | — | — | — |
| 252 | Qwen 2.5 (max) (partial coverage) | Qwen | 47.3 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1403 | — | — | — |
| 253 | Hunyuan T1 (partial coverage) | Tencent | 47.2 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1399 | — | — | — |
| 254 | GPT-4.1 | OpenAI | 47.2 | 3 of 15 | $2.00 / $8.00 | 48.5% | — | — | 31.1% | — | — | — | — | — | — | — | 1456 | — | — | — |
| 255 | Gemini 2.5 Flash Lite (nothinking) (partial coverage) | Google | 47.0 | 1 of 15 | $0.10 / $0.40 | — | — | — | — | — | — | — | — | — | — | — | 1397 | — | — | — |
| 256 | GPT-5.6 Sol (low) (partial coverage) | OpenAI | 47.0 | 2 of 15 | $5.00 / $30.00 | — | — | — | — | — | 45.4% | — | 35.4% | — | — | — | — | — | — | — |
| 257 | Devstral Small 2505 (partial coverage) | Mistral AI | 46.9 | 1 of 15 | — | 46.8% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 258 | Nova 2 Lite (partial coverage) | Amazon | 46.9 | 1 of 15 | $0.30 / $2.50 | — | — | — | — | — | — | — | — | — | — | — | 1395 | — | — | — |
| 259 | Laguna M.1 (partial coverage) | Poolside | 46.9 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | — | 1347 | — | — |
| 260 | Hunyuan TurboS (partial coverage) | Tencent | 46.8 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1394 | — | — | — |
| 261 | DeepSeek V3.2 Exp (partial coverage) | DeepSeek | 46.7 | 2 of 15 | $0.27 / $0.41 | — | — | — | — | — | — | — | — | — | — | — | 1465 | 1272 | — | — |
| 262 | Composer 2.5 (none) (partial coverage) | Cursor | 46.6 | 1 of 15 | — | — | — | — | — | — | — | — | 25.6% | — | — | — | — | — | — | — |
| 263 | Llama 3.1 Nemotron Ultra 253B v1 (partial coverage) | NVIDIA | 46.5 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1391 | — | — | — |
| 264 | Ring Flash 2.0 (partial coverage) | inclusionAI | 46.4 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1390 | — | — | — |
| 265 | GPT-5.2-Codex | OpenAI | 46.3 | 3 of 15 | $1.75 / $14.00 | 72.8% | — | — | — | 66.3% | — | — | — | — | — | — | — | 1338 | — | — |
| 266 | Kimi K2 Instruct (partial coverage) | Moonshot AI | 46.2 | 1 of 15 | — | 43.8% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 267 | GLM 5.2 (none) (partial coverage) | Z.ai | 46.2 | 1 of 15 | $0.462 / $1.452 | — | — | — | — | — | — | — | 24.5% | — | — | — | — | — | — | — |
| 268 | Command A (03-2025) (partial coverage) | Cohere | 46.0 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1390 | — | — | — |
| 269 | o3 Mini | OpenAI | 45.8 | 4 of 15 | $1.10 / $4.40 | 42.4% | 32.3% | — | — | — | — | 63.0% | — | — | — | — | 1416 | — | — | — |
| 270 | Mimo v2 Flash (partial coverage) | Xiaomi | 45.8 | 2 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1446 | 1330 | — | — |
| 271 | Claude Sonnet 4.6 (max) (partial coverage) | Anthropic | 45.8 | 1 of 15 | $3.00 / $15.00 | — | — | — | — | — | — | — | 24.3% | — | — | — | — | — | — | — |
| 272 | Magistral Medium 2506 (partial coverage) | Mistral AI | 45.7 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1387 | — | — | — |
| 273 | GPT-5.1-Codex (partial coverage) | OpenAI | 45.7 | 1 of 15 | $1.25 / $10.00 | — | — | — | — | — | — | — | — | — | — | — | — | 1336 | — | — |
| 274 | Nova Premier 1.0 (partial coverage) | Amazon | 45.6 | 1 of 15 | $2.50 / $12.50 | 42.4% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 275 | O1 Mini (partial coverage) | OpenAI | 45.5 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1387 | — | — | — |
| 276 | Grok 3 Mini Beta (partial coverage) | xAI | 45.4 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1386 | — | — | — |
| 277 | Qwen3 30B A3B (partial coverage) | Qwen | 45.3 | 1 of 15 | $0.12 / $0.50 | — | — | — | — | — | — | — | — | — | — | — | 1386 | — | — | — |
| 278 | Grok 4.6 (low) (partial coverage) | SpaceXAI | 45.2 | 1 of 15 | $2.00 / $6.00 | — | — | — | — | — | 41.6% | — | — | — | — | — | — | — | — | — |
| 279 | Mistral Large 3 (partial coverage) | Mistral AI | 45.0 | 2 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1468 | 1230 | — | — |
| 280 | GLM 4.6 | Z.ai | 44.9 | 3 of 15 | $0.55 / $2.20 | 55.4% | — | — | — | — | — | — | — | — | — | — | 1458 | 1340 | — | — |
| 281 | Claude Opus 4.6 (medium) (partial coverage) | Anthropic | 44.9 | 1 of 15 | $5.00 / $25.00 | — | — | — | — | — | — | — | 23.7% | — | — | — | — | — | — | — |
| 282 | QwQ 32B (partial coverage) | Qwen | 44.9 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1384 | — | — | — |
| 283 | Grok 4.1 Fast (reasoning) (partial coverage) | xAI | 44.8 | 2 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1461 | 1240 | — | — |
| 284 | GPT-5 Nano (high) (partial coverage) | OpenAI | 44.7 | 1 of 15 | $0.05 / $0.40 | — | — | — | — | — | — | — | — | — | — | — | 1384 | — | — | — |
| 285 | Gemini 2.5 Flash Lite (thinking) (partial coverage) | Google | 44.6 | 1 of 15 | $0.10 / $0.40 | — | — | — | — | — | — | — | — | — | — | — | 1384 | — | — | — |
| 286 | Claude Sonnet 5 (medium) (partial coverage) | Anthropic | 44.6 | 2 of 15 | $2.00 / $10.00 | — | — | — | — | — | 39.8% | — | 35.2% | — | — | — | — | — | — | — |
| 287 | Gemini 3.1 Flash Lite Preview (partial coverage) | Google | 44.5 | 2 of 15 | $0.25 / $1.50 | — | — | — | — | — | — | — | — | — | — | — | 1457 | 1254 | — | — |
| 288 | OLMo 3.1 32B Instruct (partial coverage) | Allen Institute for AI | 44.4 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1382 | — | — | — |
| 289 | Claude Opus 4.8 (low) (partial coverage) | Anthropic | 44.3 | 2 of 15 | $5.00 / $25.00 | — | — | — | — | — | 40.8% | — | 35.1% | — | — | — | — | — | — | — |
| 290 | GPT-4.1 Nano (partial coverage) | OpenAI | 44.3 | 1 of 15 | $0.10 / $0.40 | — | — | — | — | — | — | — | — | — | — | — | 1374 | — | — | — |
| 291 | Laguna XS.2 (partial coverage) | Poolside | 44.2 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | — | 1303 | — | — |
| 292 | gpt-oss-20b (partial coverage) | OpenAI | 44.0 | 1 of 15 | $0.03 / $0.13 | — | — | — | — | — | — | — | — | — | — | — | 1370 | — | — | — |
| 293 | Devstral Small 2507 (partial coverage) | Mistral AI | 43.9 | 1 of 15 | — | 38.0% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 294 | DeepSeek V4 Flash 0423 (xhigh) (partial coverage) | DeepSeek | 43.8 | 2 of 15 | $0.0643 / $0.1285 | — | — | — | — | — | — | — | — | — | — | — | — | — | 20.9% | 27.0 |
| 295 | Mercury (partial coverage) | Inception | 43.8 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1367 | — | — | — |
| 296 | Gemini 3.6 Flash (low) (partial coverage) | Google | 43.6 | 1 of 15 | $0.75 / $3.75 | — | — | — | — | — | — | — | 22.8% | — | — | — | — | — | — | — |
| 297 | GPT-5 Nano (medium) (partial coverage) | OpenAI | 43.5 | 1 of 15 | $0.05 / $0.40 | 34.8% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 298 | OLMo 3 32B Think (partial coverage) | Allen Institute for AI | 43.5 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1364 | — | — | — |
| 299 | Claude Haiku 4.5 | Anthropic | 43.4 | 3 of 15 | $1.00 / $5.00 | — | — | — | — | 64.7% | — | — | — | — | — | — | 1479 | 1326 | — | — |
| 300 | Llama 3.3 Nemotron Super 49B v1 (partial coverage) | NVIDIA | 43.3 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1363 | — | — | — |
| 301 | Nemotron 3 Nano 30B A3B (partial coverage) | NVIDIA | 43.2 | 1 of 15 | $0.05 / $0.20 | — | — | — | — | — | — | — | — | — | — | — | 1362 | — | — | — |
| 302 | GPT-4o (2024-11-20) (partial coverage) | OpenAI | 43.2 | 1 of 15 | $2.50 / $10.00 | 31.0% | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 303 | Claude Sonnet 4.6 (medium) (partial coverage) | Anthropic | 43.1 | 1 of 15 | $3.00 / $15.00 | — | — | — | — | — | — | — | 21.1% | — | — | — | — | — | — | — |
| 304 | Mistral Small 3.1 24B Instruct 2503 (partial coverage) | Mistral AI | 42.9 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1362 | — | — | — |
| 305 | GPT-5.4 Mini (medium) (partial coverage) | OpenAI | 42.7 | 1 of 15 | $0.75 / $4.50 | — | — | — | — | — | — | — | 20.8% | — | — | — | — | — | — | — |
| 306 | Gemma 3 27B (partial coverage) | Google | 42.7 | 1 of 15 | $0.08 / $0.45 | — | — | — | — | — | — | — | — | — | — | — | 1358 | — | — | — |
| 307 | Gemini 1.5 Pro 002 (partial coverage) | Google | 42.5 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1356 | — | — | — |
| 308 | Hunyuan Large Vision (partial coverage) | Tencent | 42.4 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1356 | — | — | — |
| 309 | KAT-Coder-Pro V1 (partial coverage) | Kwaipilot | 42.4 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | — | 1255 | — | — |
| 310 | Qwen2.5 72B Instruct (partial coverage) | Qwen | 42.3 | 1 of 15 | $0.36 / $0.40 | — | — | — | — | — | — | — | — | — | — | — | 1356 | — | — | — |
| 311 | DeepSeek V4 Flash 0731 (high) (partial coverage) | DeepSeek | 42.2 | 1 of 15 | $0.14 / $0.28 | — | — | — | — | — | — | — | 18.8% | — | — | — | — | — | — | — |
| 312 | Mistral Large 2407 (partial coverage) | Mistral | 42.0 | 1 of 15 | $2.00 / $6.00 | — | — | — | — | — | — | — | — | — | — | — | 1354 | — | — | — |
| 313 | Mimo v2 Flash (thinking) (partial coverage) | Xiaomi | 41.9 | 2 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1431 | 1292 | — | — |
| 314 | GPT-4o (2024-05-13) | OpenAI | 41.8 | 4 of 15 | $5.00 / $15.00 | 38.8% | 31.3% | 12.0% | — | — | — | — | — | — | — | — | 1369 | — | — | — |
| 315 | Claude Opus 4.6 (low) (partial coverage) | Anthropic | 41.8 | 1 of 15 | $5.00 / $25.00 | — | — | — | — | — | — | — | 18.3% | — | — | — | — | — | — | — |
| 316 | Step 1o Turbo 202506 (partial coverage) | StepFun | 41.7 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1352 | — | — | — |
| 317 | GPT-5.6 Terra (medium) (partial coverage) | OpenAI | 41.6 | 2 of 15 | $1.00 / $6.00 | — | — | — | — | — | 35.1% | — | 33.9% | — | — | — | — | — | — | — |
| 318 | Qwen3.5-35B-A3B (partial coverage) | Qwen | 41.5 | 2 of 15 | $0.225 / $1.80 | — | — | — | — | — | — | — | — | — | — | — | 1435 | 1250 | — | — |
| 319 | GPT-5.1-Codex-Mini (partial coverage) | OpenAI | 41.5 | 1 of 15 | $0.25 / $2.00 | — | — | — | — | — | — | — | — | — | — | — | — | 1244 | — | — |
| 320 | Gemini 2.5 Flash | Google | 41.4 | 3 of 15 | $0.30 / $2.50 | 28.7% | — | — | — | — | — | 61.9% | — | — | — | — | 1424 | — | — | — |
| 321 | DeepSeek V4 Pro (none) (partial coverage) | DeepSeek | 41.4 | 1 of 15 | $1.168 / $2.336 | — | — | — | — | — | — | — | 17.6% | — | — | — | — | — | — | — |
| 322 | Gemini 1.5 Pro 001 (partial coverage) | Google | 41.3 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1347 | — | — | — |
| 323 | Mistral Large 2411 (partial coverage) | Mistral AI | 41.1 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1346 | — | — | — |
| 324 | DeepSeek V3 | DeepSeek | 41.1 | 3 of 15 | $0.2574 / $1.0287 | — | 36.7% | — | — | — | — | 27.2% | — | — | — | — | 1388 | — | — | — |
| 325 | GPT-4.1 Mini (partial coverage) | OpenAI | 41.1 | 2 of 15 | $0.40 / $1.60 | 23.9% | — | — | — | — | — | — | — | — | — | — | 1433 | — | — | — |
| 326 | Llama 3.3 70B Instruct (partial coverage) | Meta | 41.0 | 1 of 15 | $0.10 / $0.32 | — | — | — | — | — | — | — | — | — | — | — | 1346 | — | — | — |
| 327 | Qwen3.5-Flash (partial coverage) | Qwen | 40.9 | 2 of 15 | $0.065 / $0.26 | — | — | — | — | — | — | — | — | — | — | — | 1437 | 1238 | — | — |
| 328 | Amazon Nova Pro v1.0 (partial coverage) | Amazon | 40.9 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1343 | — | — | — |
| 329 | Gemini 2.0 Flash Lite Preview 02 05 (partial coverage) | Google | 40.7 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1343 | — | — | — |
| 330 | Qwen 2.5 (partial coverage) | Qwen | 40.6 | 2 of 15 | — | 40.2% | 24.7% | — | — | — | — | — | — | — | — | — | — | — | — | — |
| 331 | DeepSeek V3.2 | DeepSeek | 40.6 | 3 of 15 | $0.269 / $0.40 | — | — | — | — | 59.0% | — | — | — | — | — | — | 1470 | 1325 | — | — |
| 332 | Claude Opus 4.8 (medium) | Anthropic | 40.6 | 4 of 15 | $5.00 / $25.00 | — | — | — | — | — | 48.7% | — | 40.5% | — | — | — | — | — | 20.9% | 19.0 |
| 333 | Claude Sonnet 4.6 (low) (partial coverage) | Anthropic | 40.5 | 1 of 15 | $3.00 / $15.00 | — | — | — | — | — | — | — | 15.0% | — | — | — | — | — | — | — |
| 334 | OLMo 3.1 32B Think (partial coverage) | Allen Institute for AI | 40.3 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1338 | — | — | — |
| 335 | Llama 3.1 70B Instruct (partial coverage) | Meta | 40.2 | 1 of 15 | $0.40 / $0.40 | — | — | — | — | — | — | — | — | — | — | — | 1333 | — | — | — |
| 336 | Claude 3 Sonnet (partial coverage) | Anthropic | 40.1 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1318 | — | — | — |
| 337 | MiniMax M3 (none) (partial coverage) | MiniMax | 40.1 | 1 of 15 | $0.30 / $1.20 | — | — | — | — | — | — | — | 14.7% | — | — | — | — | — | — | — |
| 338 | Gemma 3 12B (partial coverage) | Google | 39.9 | 1 of 15 | $0.05 / $0.15 | — | — | — | — | — | — | — | — | — | — | — | 1316 | — | — | — |
| 339 | Gemini 1.5 Flash 002 (partial coverage) | Google | 39.8 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1316 | — | — | — |
| 340 | Mistral Small 24B Instruct 2501 (partial coverage) | Mistral AI | 39.6 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1312 | — | — | — |
| 341 | Gemini 1.5 Flash 001 (partial coverage) | Google | 39.5 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1309 | — | — | — |
| 342 | Gemma 3n E4B (partial coverage) | Google | 39.4 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1308 | — | — | — |
| 343 | MiniMax M2 | MiniMax | 39.2 | 3 of 15 | $0.255 / $1.02 | 61.0% | — | — | — | — | — | — | — | — | — | — | 1385 | 1297 | — | — |
| 344 | Amazon Nova Lite v1.0 (partial coverage) | Amazon | 39.2 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1306 | — | — | — |
| 345 | Qwen3.7 Plus (none) (partial coverage) | Qwen | 39.2 | 1 of 15 | $0.32 / $1.28 | — | — | — | — | — | — | — | 10.2% | — | — | — | — | — | — | — |
| 346 | Claude Sonnet 5 (low) (partial coverage) | Anthropic | 39.2 | 2 of 15 | $2.00 / $10.00 | — | — | — | — | — | 30.5% | — | 28.7% | — | — | — | — | — | — | — |
| 347 | GPT 4 1106 Preview | OpenAI | 39.1 | 4 of 15 | — | 22.4% | 28.3% | 12.5% | — | — | — | — | — | — | — | — | 1340 | — | — | — |
| 348 | Grok 4 Fast (reasoning) (partial coverage) | xAI | 39.0 | 2 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1436 | 1161 | — | — |
| 349 | Amazon Nova Micro v1.0 (partial coverage) | Amazon | 39.0 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1289 | — | — | — |
| 350 | Command R (08-2024) (partial coverage) | Cohere | 38.8 | 1 of 15 | $0.15 / $0.60 | — | — | — | — | — | — | — | — | — | — | — | 1281 | — | — | — |
| 351 | SWE-1.6 (none) (partial coverage) | Cognition | 38.7 | 1 of 15 | — | — | — | — | — | — | — | — | 9.4% | — | — | — | — | — | — | — |
| 352 | Command R+ (08-2024) (partial coverage) | Cohere | 38.7 | 1 of 15 | $2.50 / $10.00 | — | — | — | — | — | — | — | — | — | — | — | 1280 | — | — | — |
| 353 | Trinity Large Thinking (partial coverage) | Arcee AI | 38.7 | 2 of 15 | $0.22 / $0.85 | — | — | — | — | — | — | — | — | — | — | — | 1414 | 1239 | — | — |
| 354 | OLMo 2 0325 32B Instruct (partial coverage) | Allen Institute for AI | 38.5 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1280 | — | — | — |
| 355 | Grok Code Fast 1 (partial coverage) | xAI | 38.5 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | — | 1164 | — | — |
| 356 | Mixtral 8x22B Instruct (partial coverage) | Mistral | 38.4 | 1 of 15 | $2.00 / $6.00 | — | — | — | — | — | — | — | — | — | — | — | 1277 | — | — | — |
| 357 | GPT-5.4 Mini (low) (partial coverage) | OpenAI | 38.3 | 1 of 15 | $0.75 / $4.50 | — | — | — | — | — | — | — | 9.3% | — | — | — | — | — | — | — |
| 358 | Gemma 3 4B (partial coverage) | Google | 38.3 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1274 | — | — | — |
| 359 | Gemini 1.5 Flash 8B 001 (partial coverage) | Google | 38.1 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1272 | — | — | — |
| 360 | Llama 3.1 8B Instruct (partial coverage) | Meta | 38.0 | 1 of 15 | $0.05 / $0.08 | — | — | — | — | — | — | — | — | — | — | — | 1260 | — | — | — |
| 361 | Devstral Medium 2507 (partial coverage) | Mistral AI | 37.9 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | — | 1080 | — | — |
| 362 | Mistral Medium 3.5 (none) (partial coverage) | Mistral | 37.9 | 1 of 15 | $1.50 / $7.50 | — | — | — | — | — | — | — | 8.0% | — | — | — | — | — | — | — |
| 363 | GPT-4o (partial coverage) | OpenAI | 37.9 | 1 of 15 | $2.50 / $10.00 | — | — | — | 12.2% | — | — | — | — | — | — | — | — | — | — | — |
| 364 | QwQ 32B Preview (partial coverage) | Qwen | 37.9 | 1 of 15 | — | — | — | — | — | — | — | — | — | — | — | — | 1173 | — | — | — |
| 365 | gpt-oss-120b (partial coverage) | OpenAI | 37.8 | 2 of 15 | $0.03 / $0.17 | 26.0% | — | — | — | — | — | — | — | — | — | — | 1390 | — | — | — |
| 366 | Devstral 2 (partial coverage) | Mistral AI | 37.2 | 2 of 15 | — | 53.8% | — | — | — | — | — | — | — | — | — | — | — | 1194 | — | — |
| 367 | GPT-5 Mini (partial coverage) | OpenAI | 36.9 | 2 of 15 | $0.25 / $2.00 | 56.2% | — | — | — | 39.7% | — | — | — | — | — | — | — | — | — | — |
| 368 | GPT-4o (2024-08-06) | OpenAI | 36.9 | 5 of 15 | $2.50 / $10.00 | 27.0% | 39.7% | — | 30.4% | — | — | 29.5% | — | — | — | — | 1360 | — | — | — |
| 369 | GPT-5.6 Luna (medium) (partial coverage) | OpenAI | 35.7 | 2 of 15 | $0.10 / $0.60 | — | — | — | — | — | 11.3% | — | 25.7% | — | — | — | — | — | — | — |
| 370 | Mercury 2 (partial coverage) | Inception | 35.6 | 2 of 15 | $0.25 / $0.75 | — | — | — | — | — | — | — | — | — | — | — | 1394 | 1166 | — | — |
| 371 | Claude Sonnet 4.6 (high) (partial coverage) | Anthropic | 35.4 | 2 of 15 | $3.00 / $15.00 | — | — | — | — | — | 29.9% | — | 23.5% | — | — | — | — | — | — | — |
| 372 | Llama 4 Maverick 17B 128e Instruct (partial coverage) | Meta | 35.4 | 2 of 15 | — | 21.0% | — | — | — | — | — | — | — | — | — | — | 1373 | — | — | — |
| 373 | GPT-5.6 Terra (low) (partial coverage) | OpenAI | 35.3 | 2 of 15 | $1.00 / $6.00 | — | — | — | — | — | 24.1% | — | 24.1% | — | — | — | — | — | — | — |
| 374 | Claude 3 Opus | Anthropic | 35.0 | 4 of 15 | — | 15.8% | 26.3% | 10.5% | — | — | — | — | — | — | — | — | 1355 | — | — | — |
| 375 | Gemini 2.0 Flash 001 (partial coverage) | Google | 34.4 | 2 of 15 | — | 13.5% | — | — | — | — | — | — | — | — | — | — | 1365 | — | — | — |
| 376 | GPT-4 Turbo (partial coverage) | OpenAI | 34.1 | 2 of 15 | $10.00 / $30.00 | — | — | — | — | — | — | 28.7% | — | — | — | — | 1347 | — | — | — |
| 377 | Llama 4 Scout 17B 16e Instruct (partial coverage) | Meta | 33.7 | 2 of 15 | — | 9.1% | — | — | — | — | — | — | — | — | — | — | 1362 | — | — | — |
| 378 | Claude 2 | Anthropic | 33.7 | 3 of 15 | — | 4.4% | 3.0% | 2.0% | — | — | — | — | — | — | — | — | — | — | — | — |
| 379 | GPT-4o-mini (2024-07-18) (partial coverage) | OpenAI | 33.2 | 2 of 15 | $0.15 / $0.60 | — | — | — | — | — | — | 27.5% | — | — | — | — | 1349 | — | — | — |
| 380 | Granite 4.1 8B (partial coverage) | IBM | 32.3 | 2 of 15 | $0.05 / $0.10 | — | — | — | — | — | — | — | — | — | — | — | 1353 | 1192 | — | — |
| 381 | Qwen2.5 Coder 32B Instruct (partial coverage) | Qwen | 31.6 | 2 of 15 | — | 9.0% | — | — | — | — | — | — | — | — | — | — | 1342 | — | — | — |
| 382 | GPT-5.6 Luna (low) (partial coverage) | OpenAI | 30.7 | 2 of 15 | $0.10 / $0.60 | — | — | — | — | — | 1.6% | — | 15.4% | — | — | — | — | — | — | — |
| 383 | Claude Opus 4.7 (medium) | Anthropic | 29.6 | 3 of 15 | $5.00 / $25.00 | — | — | — | — | — | — | — | 28.9% | — | — | — | — | — | 7.0% | 7.0 |
| 384 | Gemini 3.1 Pro Preview (high) (partial coverage) | Google | 29.2 | 2 of 15 | $2.00 / $12.00 | — | — | — | — | — | 11.7% | — | — | — | 8.9% | — | — | — | — | — |
| 385 | GLM 5.2 (high) | Z.ai | 29.2 | 3 of 15 | $0.462 / $1.452 | — | — | — | — | — | 36.3% | — | — | — | — | — | — | — | 17.4% | 18.0 |
| 386 | SWE Llama | Princeton NLP | 29.0 | 3 of 15 | — | 1.4% | 1.3% | 0.7% | — | — | — | — | — | — | — | — | — | — | — | — |
| 387 | Claude 3 Haiku | Anthropic | 27.8 | 3 of 15 | $0.25 / $1.25 | 40.6% | — | — | — | — | — | 20.2% | — | — | — | — | 1301 | — | — | — |
| 388 | SWE Llama 13B | Princeton NLP | 27.4 | 3 of 15 | — | 1.2% | 1.0% | 0.7% | — | — | — | — | — | — | — | — | — | — | — | — |
| 389 | GPT 3.5 | OpenAI | 22.7 | 3 of 15 | — | 0.4% | 0.3% | 0.2% | — | — | — | — | — | — | — | — | — | — | — | — |

## How this ranks

Every benchmark value becomes a percentile among the models that have it, so accuracy scores, Elo ratings and word error rates compare without hand-tuned scaling. Metrics where lower is better are inverted first. Raw values are never summed or averaged across benchmarks. A model's mean percentile is then shrunk toward the mean of the models that were broadly benchmarked, so a model tested twice cannot outrank a broadly tested one on two lucky results. Turning a data source off runs that same ranking code again in your browser over the sources you left on.

A model scored on fewer than 3 of the 15 ranked benchmarks in this category still ranks here, on the benchmarks it does have, and its row carries a partial coverage mark. On an equal score it sits under the model that earned the same number across more of the board.

## Data sources

- [DeepSWE v1.1](https://deepswe.datacurve.ai/): Pass@1 across all 113 DeepSWE v1.1 tasks, run by Datacurve with mini-swe-agent through Pier; reasoning-effort configurations remain separate.
- [Epoch AI Benchmarking Hub](https://epoch.ai/benchmarks): CC BY 4.0. Benchmark runs by Epoch AI, from the AI Benchmarking Hub.
- [FrontierCode 1.1](https://cognition.com/frontiercode): Main weighted rubric Scores on FrontierCode 1.1's 100 hardest tasks, run by Cognition across five trials; results identify model, harness, and effort.
- [FrontierSWE](https://www.frontierswe.com/): Mean@5 Dominance across FrontierSWE's complete 17-task cohort; results identify the evaluated model and harness.
- [LiveCodeBench](https://livecodebench.github.io/leaderboard): MIT. Maintainer-published Code Generation pass@1 over the 454-problem release_v6 window (2024-08-01 through 2025-05-01), from the stable 2025-08-01 snapshot; it does not cover newer frontier families.
- [LMArena](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset): CC BY 4.0. Arena ratings by LMArena, from the public leaderboard dataset.
- [SWE-bench](https://www.swebench.com): Resolve rates published by the SWE-bench maintainers.
- [Warden](https://warden.sentry.dev/benchmarking): Security review results published by Warden.