BenchLeader
Menu
Best LLMsChoose a modelLeaderboardsModelsBenchmarksProvidersReleases9

LiveBench 1.0 Coding

Coding category of LiveBench 1.0, frozen at its final release (25 June 2026). LiveBench 2.0's Coding suite is a different test.

As of 25 Oct 2026, Claude Sonnet 5.5 leads LiveBench 1.0 Coding on BenchLeader with 91.4%, ahead of Claude Opus 5.5 at 89.3%, across 67 model configurations with a published result.

Published by
LiveBench
Category
Coding
Index weight
Reference only
Models
67
Data as of
25 Oct 2026
Newest result
25 Jun 2026

The publisher has posted no new results on this board since 25 Jun 2026. BenchLeader re-reads it every day, so newer models appear here as soon as they are published.

LiveBench (livebench.ai), White et al. 2024.

This board is frozen. A different, newer test now carries its name, and its scores are on LiveBench 2.0 Coding.

What the test looks like

LiveBench 1.0's coding tasks: completing and generating code for recent competitive-programming problems, checked by running tests.

How it is scored

Percent of problems passing tests, from LiveBench 1.0's final release (25 June 2026). LiveBench 2.0's Coding suite is a different test, so these scores no longer update.

What to keep in mind

Short, self-contained problems, so it measures algorithmic coding rather than working in a codebase. See LiveBench 2.0 Coding and SWE-bench for that.

Top 15
  1. 1Claude Sonnet 5.5 (max)91.4%
  2. 2Claude Opus 5.5 (max)89.3%
  3. 3Claude Opus 5.5 (xhigh)89.3%
  4. 4Claude Sonnet 5.5 (xhigh)88.9%
  5. 5Claude Fable 5.1 (max)86.4%
  6. 6Claude Fable 5 (max)86.0%
  7. 7GPT-5.6 Sol (max)83.9%
  8. 8GPT-5.2-Codex83.6%
  9. 9GPT-5.6 Luna (max)82.9%
  10. 10GPT-5.5 (xhigh)82.2%
  11. 11Claude Opus 4.7 (xhigh)82.1%
  12. 12Claude Opus 4.8 (max)81.8%
  13. 13GPT-6 Sol (max)81.8%
  14. 14Claude Opus 5 (max)81.5%
  15. 15Kimi K381.5%
67 of 67
#
1Claude Sonnet 5.5max91.4%69.42026-06-25
2Claude Opus 5.5max89.3%72.62026-06-25
3Claude Opus 5.5xhigh89.3%70.02026-06-25
4Claude Sonnet 5.5xhigh88.9%67.92026-06-25
5Claude Fable 5.1max86.4%70.52026-06-25
6Claude Fable 5max86.0%69.32026-06-25
7GPT-5.6 Solmax83.9%68.02026-06-25
8GPT-5.2-Codex83.6%57.32026-06-25
9GPT-5.6 Lunamax82.9%59.92026-06-25
10GPT-5.5xhigh82.2%66.12026-06-25
11Claude Opus 4.7xhigh82.1%57.42026-06-25
12Claude Opus 4.8max81.8%63.32026-06-25
13GPT-6 Solmax81.8%66.42026-06-25
14Claude Opus 5max81.5%68.82026-06-25
15Kimi K381.5%58.82026-06-25
16Muse Spark 1.3xhigh81.1%64.92026-06-25
17GPT-6.1 Solxhigh80.7%67.72026-06-25
18Muse Wondermax80.7%67.52026-06-25
19Claude Sonnet 5xhigh80.7%59.02026-06-25
20GPT-6 Astramax80.4%70.42026-06-25
21GPT-6.1 Solmax80.4%69.92026-06-25
22DeepSeek V4.1 Flashmax80.0%61.22026-06-25
23Claude Opus 4.5high79.7%59.22026-06-25
24GLM 5.279.7%51.82026-06-25
25Claude Sonnet 4.6medium79.3%–2026-06-25
26GLM 5.3 Flash79.0%62.72026-06-25
27GPT-6 Lunamax79.0%60.02026-06-25
28GLM 5.379.0%–2026-06-25
29Gemini 3.7 Flashhigh78.9%63.52026-06-25
30Kimi K2.6thinking78.6%59.42026-06-25
31GPT-5.6 Terramax78.3%64.42026-06-25
32Gemini 3.5 Flashhigh78.2%63.12026-06-25
33Claude Opus 4.6high78.2%62.72026-06-25
34Qwen3.6 Plus78.2%58.82026-06-25
35Gemini 3.6 Flashhigh77.9%62.02026-06-25
36Claude Haiku 5.5max77.9%60.52026-06-25
37GPT-5.4xhigh77.5%64.72026-06-25
38Muse Spark 1.2xhigh77.5%63.22026-06-25
39Grok 4.7xhigh77.2%61.62026-06-25
40Muse Spark 1.1xhigh77.2%61.62026-06-25
41DeepSeek V4 Pro77.2%53.42026-06-25
42Mistral Large 4high77.2%49.82026-06-25
43Smaug Flash77.1%–2026-06-25
44Grok 4.676.8%57.82026-06-25
45Gemini 3.1 Prohigh76.5%60.32026-06-25
46Claude Haiku 5.5xhigh76.4%61.52026-06-25
47GPT-5.2high76.1%56.02026-06-25
48Gemini 3.5 Flash Litehigh76.1%47.92026-06-25
49Ox Alphamax75.8%–2026-06-25
50Qwen3.8 27B75.7%54.62026-06-25
51Smaug Mini75.4%–2026-06-25
52DeepSeek V4 Flash75.0%52.22026-06-25
53Qwen3.7max74.2%60.62026-06-25
54Kimi K2.7 Code74.0%55.82026-06-25
55Qwen3.8 Maxmax72.9%63.32026-06-25
56Qwen3.8-Flash-Next72.5%60.32026-06-25
57Gemini 3.8 Flashhigh72.5%64.42026-06-25
58Qwen3.6 27B71.8%48.32026-06-25
59GPT-5.4 Minixhigh71.6%56.32026-06-25
60GPT-5.4 Nanoxhigh70.8%54.22026-06-25

More coding benchmarks · All LLM benchmarks

Cite as: BenchLeader, “LiveBench 1.0 Coding leaderboard”, data as of 25 Oct 2026, https://www.benchleader.com/benchmarks/livebench_coding.

LiveBench 1.0 Coding: questions

What does LiveBench 1.0 Coding measure?
LiveBench 1.0's coding tasks: completing and generating code for recent competitive-programming problems, checked by running tests. Scores are reported in percent of tasks solved; higher is better.
Which AI model leads LiveBench 1.0 Coding?
Claude Sonnet 5.5 leads LiveBench 1.0 Coding with 91.4% as of 25 Oct 2026, ahead of Claude Opus 5.5 at 89.3%.
How many models have LiveBench 1.0 Coding results?
67 model configurations have a LiveBench 1.0 Coding result on BenchLeader, all taken from LiveBench.
Who runs LiveBench 1.0 Coding and how often is it updated?
LiveBench 1.0 Coding is published by LiveBench. BenchLeader re-reads the published results once a day and records the date each result was published.
Does LiveBench 1.0 Coding count toward the BenchLeader Index?
No. LiveBench 1.0 Coding is shown for reference but left out of the composite index.