Skip to main content
BenchLM
Data

Artificial Analysis LiveCodeBench (AA LiveCodeBench)

We show this table for reference; we do not rank on it.

Data verified

An independently evaluated LiveCodeBench result from Artificial Analysis.

Benchmark score on AA LiveCodeBench — November 13, 2026

We compile the AA LiveCodeBench rows from secondary reports. Gemini 3 Pro leads the table at 91.7%, followed by GLM-4.7 (89.4%) and GPT-OSS 120B (87.8%). We do not use these results to rank models overall.

1Closed

Gemini 3 Pro

Google

Carried from an upstream page
91.7%
Overall 58.35Context 2M
2Open weights

GLM-4.7

Z.AI

Carried from an upstream page
89.4%
Overall 47.74Context 200K
3Open weights

GPT-OSS 120B

OpenAI

Carried from an upstream page
87.8%
Overall 31.91Context 128K

13 modelsCodingCurrentDisplay onlyUpdated November 13, 2026

Benchmark score table (13 models)

None of the 13 results on this page has a documented run by the benchmark or an independent evaluator: 13 carried from an upstream page, where who ran them is not recorded.

Benchmark score results for AA LiveCodeBench
RankModel / configurationScoreParameters (B)Open / closed
1Gemini 3 ProGoogleCarried from an upstream page
91.7%
Not reportedClosed
2GLM-4.7Z.AICarried from an upstream page
89.4%
Not reportedOpen
3GPT-OSS 120BOpenAICarried from an upstream page
87.8%
Not reportedOpen
4Claude Opus 4.5 ThinkingAnthropicCarried from an upstream page
87.1%
Not reportedClosed
5GPT-5.1OpenAICarried from an upstream page
86.8%
Not reportedClosed
6o4-miniOpenAICarried from an upstream page
85.9%
Not reportedClosed
7GPT-5 (high)OpenAICarried from an upstream page
84.6%
Not reportedClosed
8GPT-5 miniOpenAICarried from an upstream page
83.8%
Not reportedClosed
9Grok 4 Fast (Reasoning)xAICarried from an upstream page
83.2%
Not reportedClosed
10Grok 4.1 Fast (Reasoning)xAICarried from an upstream page
82.2%
Not reportedClosed
11Grok 4xAICarried from an upstream page
81.9%
Not reportedClosed
12o3OpenAICarried from an upstream page
80.8%
Not reportedClosed
13Gemini 2.5 ProGoogleCarried from an upstream page
80.1%
Not reportedClosed
  • Lab-reported
  • Benchmark-run
  • Independent
What the labels mean

Among the reported AA LiveCodeBench rows, Gemini 3 Pro is first at 91.7%. The third row is 3.9 points behind. The broader top-10 range is 9.5 points, so many of the published results sit in a relatively narrow band.

13 models have been evaluated on AA LiveCodeBench. The benchmark falls in the Coding category. AA LiveCodeBench is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About AA LiveCodeBench

Year

2026

Tasks

Contamination-resistant coding tasks

Format

Pass rate

Difficulty

Competitive programming

Stored separately from the core LiveCodeBench lane because Artificial Analysis controls the evaluation configuration.

Artificial Analysis LiveCodeBench Benchmark Leaderboard

Freshness and provenance

Version

AA LiveCodeBench 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Questions

What does AA LiveCodeBench measure?

An independently evaluated LiveCodeBench result from Artificial Analysis.

Which model scores highest on AA LiveCodeBench?

Gemini 3 Pro by Google currently leads with a score of 91.7% on AA LiveCodeBench.

How many models are evaluated on AA LiveCodeBench?

13 AI models have published results on AA LiveCodeBench in the BenchLM catalog.

Compare top models on AA LiveCodeBench

Gemini 3 Pro vs GLM-4.7GLM-4.7 vs GPT-OSS 120BGPT-OSS 120B vs Claude Opus 4.5 ThinkingClaude Opus 4.5 Thinking vs GPT-5.1
Last updated: November 13, 2026 · BenchLM version AA LiveCodeBench 2026

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 7,000+ subscribers.

One email each week. Unsubscribe anytime.