FrontierCode 1.1 Main · Open Analysis

#	Model	Effort	Score	Cost / task	Harness
1	Opus 5.5	medium	54.6%	$0.802	claude-code
2	Fable 5	xhigh	53.5%	$13.09	claude-code
3	Opus 5	medium	53.4%	$4.31	claude-code
4	GPT-6 Astra	max	53.3%	$4.59	codex
5	Sonnet 5.5	xhigh	52.1%	$1.59	claude-code
6	Fable 5.1	medium	50.9%	$3.28	claude-code
7	GPT-6.1 Sol	medium	50.2%	$0.361	codex
8	SWE 2	max	50.0%	$1.18	devin
9	GPT-6 Sol	max	49.3%	$2.07	codex
10	Grok 4.6	high	48.0%	$2.88	grok-build
11	Grok 4.7	high	47.6%	$6.65	grok-build
12	Gemini 4	medium	47.5%	$6.74	chisel
13	GPT-5.6 Sol	max	47.5%	$5.19	codex
14	Opus 4.8	max	46.5%	$9.62	claude-code
15	Haiku 5.5	max	46.4%	$1.33	claude-code
16	Kimi K3	none	44.2%	$3.82	mini-swe-agent
17	DeepSeek V4.1 Flash	high	43.9%	$0.620	chisel
18	Gemini 3.7 Flash	medium	43.6%	$1.82	chisel
19	GPT-5.5	xhigh	43.0%	$4.03	codex
20	Sonnet 5	xhigh	42.7%	$10.07	claude-code
21	Grok 4.5	high	42.4%	$1.30	grok-build
22	GPT-6 Luna	max	42.4%	$0.103	codex
23	SWE 1.7	none	42.0%	$1.97	chisel
24	GPT-5.6 Terra	max	41.3%	$1.87	codex
25	Gemini 3.8 Flash	medium	41.2%	$2.60	chisel
26	GLM 5.3	max	40.1%	$16.91	chisel
27	GPT-5.6 Luna	max	39.8%	$0.370	codex
28	Mistral Large 4	high	39.4%	$4.13	chisel
29	Opus 4.7	max	38.5%	$9.09	claude-code
30	Gemini 3.6 Flash	medium	34.4%	$4.04	chisel
31	GLM 5.3 Flash	max	31.8%	$1.15	chisel
32	Kimi K2.7	none	30.1%	$3.01	mini-swe-agent
33	DeepSeek V4 Pro 0813	high	28.5%	$1.81	chisel
34	GPT-5.4 Mini	xhigh	27.0%	$1.52	codex
35	Opus 4.6	high	26.6%	$3.98	claude-code
36	Composer 2.5	none	25.6%	$3.09	cursor-cli
37	GLM 5.2	none	24.5%	$2.47	mini-swe-agent
38	Sonnet 4.6	max	24.3%	$2.90	claude-code
39	DeepSeek V4 Flash 0731	high	18.8%	$1.53	chisel
40	DeepSeek V4 Pro	none	17.6%	$1.55	mini-swe-agent
41	MiniMax M3	none	14.7%	$0.676	mini-swe-agent
42	Inkling	max	14.0%	$3.60	mini-swe-agent
43	Nemotron 3 Ultra 550B A55B	none	13.6%	$1.47	chisel
44	Qwen 3.7 Plus	none	10.2%	$0.241	mini-swe-agent
45	SWE 1.6	none	9.4%	$0.489	chisel
46	Mistral 3.5	none	8.0%	$1.35	chisel