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.

42 benchmarks4 sourcesUpdated 17 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 / 27	GLM 5.3454 GB · 23 benchmarks		131	138	129	131	126	23	454 GB
2 / 34	MiMo V2.6 Pro RL614 GB · 5 benchmarks		124	–	–	140	–	5	614 GB
3 / 36	MiMo V2.6 Flash RL187 GB · 4 benchmarks		124	–	–	122	–	4	187 GB
4 / 39	DeepSeek V4 Flash 0731183 GB · 21 benchmarks		121	–	128	124	129	21	183 GB
5 / 42	GLM 5.2454 GB · 26 benchmarks		120	128	126	129	120	26	454 GB
6 / 43	DeepSeek V4.1 Flash458 GB · 20 benchmarks		119	–	129	123	133	20	458 GB
7 / 44	Kimi K2.6618 GB · 20 benchmarks		119	122	122	122	119	20	618 GB
8 / 49	MiMo V2.5 Pro616 GB · 6 benchmarks		116	–	–	115	111	6	616 GB
9 / 50	Kimi K2.7 Code618 GB · 19 benchmarks		115	119	121	118	118	19	618 GB
10 / 51	GLM 5.1454 GB · 14 benchmarks		115	123	117	117	115	14	454 GB
11 / 54	GLM 5.3 Flash205 GB · 21 benchmarks		113	131	121	124	127	21	205 GB
12 / 55	MiniMax M3258 GB · 19 benchmarks		113	113	106	118	107	19	258 GB
13 / 56	Qwen3.8 27B17.9 GB · 12 benchmarks		112	–	118	112	124	12	17.9 GB
14 / 61	Kimi K2.5618 GB · 20 benchmarks		110	111	–	117	114	20	618 GB
15 / 62	GLM 5454 GB · 15 benchmarks		110	118	–	–	109	15	454 GB
16 / 64	MiniMax M2.5140 GB · 11 benchmarks		109	110	–	–	114	11	140 GB
17 / 66	Gemma 4 31B IT21.7 GB · 11 benchmarks		108	–	–	105	109	11	21.7 GB
18 / 67	DeepSeek R1 0528412 GB · 10 benchmarks		107	–	–	–	100	10	412 GB
19 / 68	DeepSeek V3.1 Terminus412 GB · 5 benchmarks		107	–	–	99	108	5	412 GB
20 / 69	DeepSeek V3.2 Exp413 GB · 10 benchmarks		107	–	–	101	113	10	413 GB
21 / 70	Step 3.7 Flash139 GB · 3 benchmarks	—	107	–	–	102	–	3	139 GB
22 / 71	DeepSeek V4 Flash175 GB · 10 benchmarks		107	–	–	111	112	10	175 GB
23 / 73	Qwen3.6 35B A3B22.0 GB · 12 benchmarks		105	–	109	110	107	12	22.0 GB
24 / 74	Inkling575 GB · 22 benchmarks		104	109	112	116	121	22	575 GB
25 / 75	Gemma 4 26B A4B IT16.5 GB · 9 benchmarks		104	–	–	102	101	9	16.5 GB
26 / 79	Kimi K2 Thinking617 GB · 9 benchmarks		103	108	–	–	–	9	617 GB
27 / 80	NVIDIA Nemotron 3 Ultra 550B A55B BF16378 GB · 15 benchmarks		103	–	112	112	115	15	378 GB
28 / 81	MiMo V2.5188 GB · 4 benchmarks		103	–	–	107	–	4	188 GB
29 / 83	MiniMax M2.7140 GB · 7 benchmarks		103	112	–	112	–	7	140 GB
30 / 84	Qwen3 235B A22B Thinking 2507143 GB · 13 benchmarks		102	–	–	108	106	13	143 GB
31 / 85	Qwen3.6 27B17.9 GB · 12 benchmarks		102	–	114	111	106	12	17.9 GB
32 / 89	Qwen3 235B A22B Instruct 2507143 GB · 8 benchmarks		102	–	–	–	98	8	143 GB
33 / 90	Kimi K2 Instruct617 GB · 10 benchmarks		102	102	–	–	90	10	617 GB
34 / 91	GLM 4.7218 GB · 13 benchmarks		101	106	111	111	104	13	218 GB
35 / 92	Qwen3 235B A22B143 GB · 9 benchmarks		101	–	–	–	99	9	143 GB
36 / 93	GPT OSS 20B13.1 GB · 11 benchmarks		101	–	–	92	91	11	13.1 GB
37 / 94	Qwen3 Coder 480B A35B Instruct291 GB · 6 benchmarks	—	101	105	–	–	–	6	291 GB
38 / 95	MiMo v2 Flash187 GB · 3 benchmarks	—	100	–	–	82	–	3	187 GB
39 / 96	DeepSeek R1412 GB · 13 benchmarks		100	–	–	99	96	13	412 GB
40 / 97	Kimi K2 Instruct 0905617 GB · 6 benchmarks		100	–	–	–	–	6	617 GB
41 / 98	DeepSeek v3 0324412 GB · 12 benchmarks		100	–	–	98	88	12	412 GB
42 / 101	GLM 4.5218 GB · 7 benchmarks		99	105	–	–	–	7	218 GB
43 / 104	Qwen3.5 27B17.9 GB · 6 benchmarks		98	–	–	–	111	6	17.9 GB
44 / 105	GPT OSS 120B70.8 GB · 20 benchmarks		98	82	–	103	97	20	70.8 GB
45 / 106	DeepSeek v3412 GB · 8 benchmarks		97	–	–	90	81	8	412 GB
46 / 108	GLM 4.6217 GB · 6 benchmarks		96	104	–	100	–	6	217 GB
47 / 109	NVIDIA Nemotron 3 Super 120B A12B BF1674.7 GB · 4 benchmarks		96	–	–	98	–	4	74.7 GB
48 / 111	Qwen3 Coder Next48.3 GB · 3 benchmarks	—	95	–	–	91	–	3	48.3 GB
49 / 112	Qwen3 32B22.6 GB · 9 benchmarks		93	–	–	96	91	9	22.6 GB
50 / 113	Mistral Large 3 675B Instruct 2512454 GB · 5 benchmarks		93	–	–	97	89	5	454 GB
51 / 116	Trinity Large Thinking271 GB · 3 benchmarks	—	92	–	–	97	–	3	271 GB
52 / 119	Llama 4 Maverick 17B 128E Instruct273 GB · 16 benchmarks		81	–	–	96	85	16	273 GB
53 / 120	Llama 3.3 70B Instruct51.3 GB · 12 benchmarks		76	–	–	81	82	12	51.3 GB
54 / 122	Gemma 3 27B IT21.7 GB · 10 benchmarks		70	–	–	81	74	10	21.7 GB
55 / 124	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.