Coding

LiveBench Coding Leaderboard

LiveBench Coding evaluates code generation and completion on fresh, contamination-free programming tasks that are updated regularly.

Source: livebench8 open models ranked+29 proprietaryData through Jun 2026

Open models ranked on LiveBench Coding

# shows rank among open models / rank overall (including proprietary).

#ModelScore
1 / 12GLM 5.2 · 753.3B
79.7
2 / 15Kimi K2.6 · 1058.6B
78.6
3 / 27Kimi K2.7 Code · 1058.6B
74.0
4 / 28Qwen3.6 27B · 27.8B
71.8
5 / 30Inkling · 952.4B
71.0
6 / 32DeepSeek V4 Pro · 861.6B
70.0
7 / 34DeepSeek V4 Flash · 158.1B
69.2
8 / 36MiniMax M3 · 427.0B
68.2

Score vs model size

Which models give the most quality for their size — the ones worth running locally.

100B1Tmodel size (log scale) →79.768.2Kimi K2.6 · 1.1T · 78.6Kimi K2.7 Code · 1.1T · 74.0Inkling · 952B · 71.0DeepSeek V4 Pro · 862B · 70.0DeepSeek V4 Flash · 158B · 69.2MiniMax M3 · 427B · 68.2Qwen3.6 27B · 28B · 71.8Qwen3.6 27BGLM 5.2 · 753B · 79.7GLM 5.2
Each dot is a model. Up = higher score, left = smaller (easier to run locally). The dashed line marks the efficiency frontier — the best score you can get at each size or smaller.
  • Qwen3.6 27B, 28B, score 71.8 — on the efficiency frontier (best score at its size or smaller).
  • GLM 5.2, 753B, score 79.7 — on the efficiency frontier (best score at its size or smaller).

LiveBench Coding: frequently asked questions

What is the best open LLM on LiveBench Coding?
GLM 5.2 is the top open model on LiveBench Coding, scoring 79.7. Among all models tested — including proprietary ones — it ranks #11. The top model overall is Claude Fable 5 Max (Anthropic) at 86.0.
What's the best LiveBench Coding model you can run on a 24 GB GPU?
Qwen3.6 27B is the highest-scoring open model that fits in 24 GB at 4-bit quantization (about 15 GB), scoring 71.8 on LiveBench Coding.
Can open models match proprietary models on LiveBench Coding?
Not quite on LiveBench Coding: the strongest proprietary model (Claude Fable 5 Max) scores 86.0, ahead of the best open model (GLM 5.2) at 79.7 — but you can run the open one yourself.

Scores aggregated from livebench. llmrun does not run this benchmark — see the source for methodology, or the about benchmarks for what it measures.