---
title: "Coding leaderboard"
description: "Rankings on programming prompts: writing, debugging, explaining and refactoring code."
canonical: https://pirouter.ai/rankings/coding
last-updated: 2026-10-15
---

# Coding leaderboard

> Rankings on programming prompts: writing, debugging, explaining and refactoring code.

- Updated: 2026-09-30
- 2,341,980 votes · 380 models
- Score · higher is better

## Full ranking

| Rank | Rank spread | Model | Vendor | Score | Votes |
|---|---|---|---|---|---|
| 1 | 1–12 | Gemini 4 Argon (high) | Google | 1559 ±17 | 1225 |
| 2 | 1–11 | Claude Fable 5 (high) | Anthropic | 1552 ±7 | 9946 |
| 3 | 1–10 | Claude Opus 4.6 (high) | Anthropic | 1551 ±6 | 20144 |
| 4 | 1–11 | Claude Opus 4.7 (high) | Anthropic | 1551 ±6 | 18437 |
| 5 | 1–23 | Kimi K3 (max) | Moonshot | 1541 ±8 | 7165 |
| 6 | 1–39 | MiMo V2.6 Pro | Xiaomi | 1540 ±18 | 1101 |
| 7 | 1–32 | [GPT-6 Astra \(max\)](https://pirouter.ai/models/openai/gpt-6-astra) | OpenAI | 1539 ±13 | 2010 |
| 8 | 1–27 | Muse Spark 1.3 (max) | Meta | 1539 ±11 | 3142 |
| 9 | 1–46 | [Claude Opus 5.5 \(high\)](https://pirouter.ai/models/anthropic/claude-opus-5.5) | Anthropic | 1538 ±19 | 918 |
| 10 | 1–52 | Muse Spark 1.2 (xhigh) | Meta | 1535 ±18 | 1085 |
| 11 | 5–27 | Muse Spark 1.1 | Meta | 1534 ±7 | 10381 |
| 12 | 5–29 | [GPT-5.6 Sol \(xhigh\)](https://pirouter.ai/models/openai/gpt-5.6-sol) | OpenAI | 1534 ±7 | 9788 |
| 13 | 5–27 | Claude Opus 5 (high) | Anthropic | 1534 ±6 | 15273 |
| 14 | 5–27 | Claude Opus 4.8 (high) | Anthropic | 1533 ±6 | 17275 |
| 15 | 2–44 | [Claude Fable 5.1 \(max\)](https://pirouter.ai/models/anthropic/claude-fable-5.1) | Anthropic | 1533 ±13 | 2254 |
| 16 | 5–39 | [Gemini 3.8 Flash \(high\)](https://pirouter.ai/models/google/gemini-3.8-flash) | Google | 1531 ±8 | 6660 |
| 17 | 5–38 | Claude Opus 4.5 20251101 High 32k | Anthropic | 1531 ±7 | 7790 |
| 18 | 5–44 | Muse Spark | Meta | 1530 ±10 | 3930 |
| 19 | 5–52 | [DeepSeek V4.1 Flash \(max\)](https://pirouter.ai/models/deepseek/deepseek-v4.1-flash) | DeepSeek | 1529 ±12 | 2314 |
| 20 | 5–38 | Claude Sonnet 4.6 | Anthropic | 1529 ±6 | 19706 |
| 21 | 4–68 | Qwen3.7 Preview (max) | Alibaba | 1524 ±18 | 1167 |
| 22 | 6–56 | [GLM 5.3 Flash](https://pirouter.ai/models/z-ai/glm-5.3-flash) | Z.ai | 1523 ±9 | 5833 |
| 23 | 6–49 | Claude Opus 4.5 | Anthropic | 1523 ±5 | 17972 |
| 24 | 5–65 | [GPT-6 Sol \(max\)](https://pirouter.ai/models/openai/gpt-6-sol) | OpenAI | 1523 ±14 | 1799 |
| 25 | 6–58 | Qwen3.8 (max) | Alibaba | 1522 ±8 | 6480 |
| 26 | 5–69 | MiMo V2.6 Flash | Xiaomi | 1521 ±16 | 1361 |
| 27 | 10–55 | MiMo V2.5 Pro | Xiaomi | 1521 ±6 | 19204 |
| 28 | 6–61 | GLM 5.3 (max) | Z.ai | 1521 ±10 | 4283 |
| 29 | 11–52 | Gemini 3.1 Pro Preview | Google | 1521 ±5 | 33447 |
| 30 | 11–56 | [GPT-5.4 \(high\)](https://pirouter.ai/models/openai/gpt-5.4) | OpenAI | 1521 ±6 | 17560 |
| 31 | 12–57 | Claude Sonnet 4.5 20250929 High 32k | Anthropic | 1519 ±5 | 19909 |
| 32 | 12–60 | [Claude Sonnet 5 \(high\)](https://pirouter.ai/models/anthropic/claude-sonnet-5) | Anthropic | 1519 ±6 | 12249 |
| 33 | 12–60 | [GPT-5.5 \(high\)](https://pirouter.ai/models/openai/gpt-5.5) | OpenAI | 1519 ±6 | 19151 |
| 34 | 12–61 | [Gemini 3.6 Flash \(high\)](https://pirouter.ai/models/google/gemini-3.6-flash) | Google | 1518 ±7 | 10226 |
| 35 | 11–62 | [Gemini 3.7 Flash \(high\)](https://pirouter.ai/models/google/gemini-3.7-flash) | Google | 1518 ±8 | 5873 |
| 36 | 12–62 | Gemini 3 Pro | Google | 1518 ±7 | 8788 |
| 37 | 14–62 | Kimi K2.6 | Moonshot | 1516 ±7 | 11069 |
| 38 | 16–65 | [GPT-5.6 Terra \(xhigh\)](https://pirouter.ai/models/openai/gpt-5.6-terra) | OpenAI | 1515 ±7 | 10012 |
| 39 | 16–65 | GPT-5.2 Chat Latest | OpenAI | 1515 ±7 | 9630 |
| 40 | 18–64 | Dola Seed 2.0 Pro | ByteDance | 1514 ±5 | 22007 |
| 41 | 16–68 | Qwen3.5 Preview (max) | Alibaba | 1514 ±8 | 6359 |
| 42 | 16–67 | Grok 4.5 | xAI | 1514 ±7 | 10951 |
| 43 | 16–69 | GPT-5.5 Instant | OpenAI | 1513 ±8 | 7912 |
| 44 | 19–67 | GLM 5.1 | Z.ai | 1513 ±6 | 15944 |
| 45 | 18–69 | Ernie 5.1 | Baidu | 1512 ±7 | 10904 |
| 46 | 20–66 | Claude Sonnet 4.5 | Anthropic | 1512 ±5 | 19734 |
| 47 | 19–69 | [GLM 5.2 \(max\)](https://pirouter.ai/models/z-ai/glm-5.2) | Z.ai | 1512 ±6 | 12316 |
| 48 | 10–84 | Qwen3.6 Preview (max) | Alibaba | 1512 ±15 | 1606 |
| 49 | 12–81 | [GPT-6 Luna \(max\)](https://pirouter.ai/models/openai/gpt-6-luna) | OpenAI | 1512 ±14 | 1822 |
| 50 | 19–72 | Claude Opus 4.1 (thinking) | Anthropic | 1511 ±6 | 9878 |
| 51 | 23–73 | Grok 4.20 Beta 0309 Reasoning | xAI | 1511 ±6 | 18334 |
| 52 | 26–73 | Grok 4.20 Multi Agent Beta 0309 | xAI | 1509 ±6 | 18000 |
| 53 | 23–75 | Grok 4.20 Beta1 | xAI | 1508 ±8 | 7490 |
| 54 | 28–75 | [Gemini 3.5 Flash \(high\)](https://pirouter.ai/models/google/gemini-3.5-flash) | Google | 1507 ±6 | 13673 |
| 55 | 24–80 | Grok 4.6 (high) | xAI | 1507 ±8 | 5871 |
| 56 | 20–89 | [DeepSeek V4 Pro 0813 \(high\)](https://pirouter.ai/models/deepseek/deepseek-v4-pro) | DeepSeek | 1506 ±11 | 2815 |
| 57 | 28–81 | Gemini 3 Flash | Google | 1505 ±8 | 6586 |
| 58 | 27–85 | Qwen3.8 27b | Alibaba | 1505 ±9 | 5005 |
| 59 | 32–87 | MiMo V2 Pro | Xiaomi | 1503 ±8 | 7255 |
| 60 | 23–93 | Kimi K2.5 Instant | Moonshot | 1503 ±14 | 1868 |
| 61 | 40–84 | Kimi K2.5 (thinking) | Moonshot | 1502 ±5 | 20732 |
| 62 | 20–95 | Gemma 4 31b | Google | 1502 ±15 | 1421 |
| 63 | 36–89 | [GPT-5.6 Luna \(xhigh\)](https://pirouter.ai/models/openai/gpt-5.6-luna) | OpenAI | 1502 ±7 | 10293 |
| 64 | 35–89 | Longcat Flash Chat 2602 Exp | Meituan | 1501 ±7 | 8146 |
| 65 | 30–92 | Hy3 | Tencent | 1501 ±11 | 2863 |
| 66 | 39–89 | [Gemini 3.5 Flash Lite](https://pirouter.ai/models/google/gemini-3.5-flash-lite) | Google | 1501 ±7 | 9763 |
| 67 | 42–89 | Qwen3.7 Plus | Alibaba | 1501 ±6 | 12160 |
| 68 | 48–89 | Grok 4.1 (thinking) | xAI | 1500 ±6 | 15408 |
| 69 | 48–91 | GLM 5 | Z.ai | 1498 ±7 | 7621 |
| 70 | 44–92 | Claude Opus 4 (thinking) | Anthropic | 1498 ±8 | 6557 |
| 71 | 49–91 | GPT-5.3 Chat Latest | OpenAI | 1498 ±7 | 9164 |
| 72 | 51–92 | Qwen3.6 Plus | Alibaba | 1496 ±6 | 14664 |
| 73 | 53–93 | Minimax M3 | MiniMax | 1495 ±6 | 16800 |
| 74 | 53–93 | [GPT-5.4 Mini \(high\)](https://pirouter.ai/models/openai/gpt-5.4-mini) | OpenAI | 1495 ±6 | 17208 |
| 75 | 35–110 | Grok 4.7 (xhigh) | xAI | 1492 ±17 | 1238 |
| 76 | 53–97 | GPT-5.1 (high) | OpenAI | 1492 ±7 | 8411 |
| 77 | 53–97 | Inkling | Thinky | 1492 ±7 | 9634 |
| 78 | 59–96 | Qwen3.5 397b A17b | Alibaba | 1491 ±5 | 25141 |
| 79 | 56–101 | MiMo V2.5 | Xiaomi | 1491 ±7 | 13647 |
| 80 | 58–102 | Ernie 5.0 0110 | Baidu | 1490 ±7 | 8965 |
| 81 | 60–99 | Gemini 3 Flash | Google | 1490 ±5 | 23952 |
| 82 | 60–102 | GPT-5.2 (high) | OpenAI | 1489 ±6 | 12226 |
| 83 | 56–105 | MiMo V2 Omni | Xiaomi | 1489 ±9 | 5878 |
| 84 | 53–107 | GLM 5v Turbo | Z.ai | 1488 ±11 | 2918 |
| 85 | 66–103 | Grok 4.3 | xAI | 1488 ±6 | 19933 |
| 86 | 66–105 | Kimi K2 Thinking Turbo | Moonshot | 1487 ±5 | 15272 |
| 87 | 48–122 | Amazon Nova Experimental Chat 26 02 10 | Amazon | 1486 ±19 | 879 |
| 88 | 59–114 | GLM 4.7 | Z.ai | 1484 ±12 | 2485 |
| 89 | 68–107 | [DeepSeek V4 Flash 0731](https://pirouter.ai/models/deepseek/deepseek-v4-flash) | DeepSeek | 1484 ±6 | 15253 |
| 90 | 56–119 | Gemma 4 26b A4b | Google | 1483 ±15 | 1443 |
| 91 | 78–107 | Claude Haiku 4.5 | Anthropic | 1481 ±4 | 36871 |
| 92 | 51–132 | Amazon Nova Experimental Chat 26 01 10 | Amazon | 1481 ±21 | 781 |
| 93 | 74–114 | Qwen3 Preview (max) | Alibaba | 1480 ±8 | 5336 |
| 94 | 78–111 | Minimax M2.7 | MiniMax | 1480 ±5 | 22747 |
| 95 | 54–131 | Muse Glimmer | Meta | 1480 ±18 | 1119 |
| 96 | 71–119 | Mistral Medium 3.5 | Mistral | 1479 ±11 | 3281 |
| 97 | 82–117 | DeepSeek V3.2 (thinking) | DeepSeek | 1476 ±7 | 8876 |
| 98 | 74–131 | DeepSeek V3.2 Exp (thinking) | DeepSeek | 1475 ±13 | 1916 |
| 99 | 76–130 | Nvidia Nemotron 3 Ultra 550b A55b Nvfp4 | NVIDIA | 1474 ±11 | 3195 |
| 100 | 83–122 | Claude Sonnet 4 (thinking) | Anthropic | 1474 ±8 | 6312 |

## FAQ

### What is the best AI model for coding?

As of 2026-09-30, Google Gemini 4 Argon (high) is #1 on the coding board with a score of 1559, rank spread 1–12.

### What is the best open-weights model for coding?

Kimi K3 (Moonshot), #5 on this board.

### What does this board rank?

Rankings on programming prompts: writing, debugging, explaining and refactoring code.

### How often is it updated?

Whenever new models ship or the rankings change; this board was last updated on 2026-09-30.

Data: [LMArena leaderboard dataset](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset), used under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/); filtered and reasoning tiers merged, scores unchanged.
View the original board: https://arena.ai/leaderboard/text/coding