llmrun Composite Score · Coding

The best open LLMs for coding, ranked on SWE-bench, LiveCodeBench, Aider and other coding boards. The score shown is the overall llmrun Composite Score; the highlighted column is the coding skill score used for the order.

40 benchmarks4 sourcesUpdated 10 Oct 2026Reference scale 2026-Q4How the Composite Score is built

Coding score vs VRAM needed

This chart doesn't show proprietary models.

Open models Overall

General Coding Agents Math Science Reasoning

Models ranked by coding llmrun Composite Score
# (open / overall)	Model	llmrun Composite Score	Coding	Agents	Math	Science	Reasoning	Benchmarks	VRAM
1 / 53	Qwen3.8 27B17.9 GB · 12 benchmarks		112	–	118	112	124	12	17.9 GB
2 / 64	Gemma 4 31B IT21.7 GB · 11 benchmarks		108	–	–	105	109	11	21.7 GB
3 / 70	Qwen3.6 35B A3B22.0 GB · 12 benchmarks		105	–	109	110	107	12	22.0 GB
4 / 72	Gemma 4 26B A4B IT16.5 GB · 9 benchmarks		104	–	–	102	101	9	16.5 GB
5 / 81	Qwen3.6 27B17.9 GB · 12 benchmarks		102	–	114	111	106	12	17.9 GB
6 / 89	GPT OSS 20B13.1 GB · 11 benchmarks		101	–	–	92	91	11	13.1 GB
7 / 101	Qwen3.5 27B17.9 GB · 6 benchmarks		98	–	–	–	111	6	17.9 GB
8 / 102	GPT OSS 120B70.8 GB · 20 benchmarks		98	82	–	103	97	20	70.8 GB
9 / 106	NVIDIA Nemotron 3 Super 120B A12B BF1674.7 GB · 4 benchmarks		96	–	–	98	–	4	74.7 GB
10 / 108	Qwen3 Coder Next48.3 GB · 3 benchmarks	—	95	–	–	91	–	3	48.3 GB
11 / 109	Qwen3 32B22.6 GB · 9 benchmarks		93	–	–	96	91	9	22.6 GB
12 / 117	Llama 3.3 70B Instruct51.3 GB · 12 benchmarks		76	–	–	81	82	12	51.3 GB
13 / 119	Gemma 3 27B IT21.7 GB · 10 benchmarks		69	–	–	80	74	10	21.7 GB
14 / 121	Llama 3.1 8B Instruct7.9 GB · 10 benchmarks		56	–	–	56	69	10	7.9 GB

The range after each score is a 90% interval: given the boards a model has results on, its true score is very likely inside it. Short ranges mean many agreeing results; long ranges mean few or conflicting ones.

Why some models are missing. A model receives a score only if it has results on at least 4 benchmarks across at least 2 categories, with one being reasoning or coding. Skill scores require at least 2 benchmarks for that skill; otherwise they display as –.

VRAM is llmrun's estimate at Q4_K_M (or the smallest quantization we track) for the weights plus room for a 16K-token context. Longer contexts need more memory. Proprietary models can't run locally, so the VRAM filter hides them.

How the llmrun Composite Score is built · llmrun doesn't run these benchmarks; scores come from public sources.