DeepSWE

Software engineering tasks from Datacurve, pass@1 with a fixed agent harness.

Category	Coding
Metric	Accuracy
Source	Epoch AI (evaluator)
License	CC BY 4.0
Benchmark page	deepswe.datacurve.ai
Status	Scored
Weight	1.0 within Coding
Fit	Discrimination 1.80, noise 0.33

Results

#	Model	Provider	Result	Date
1	GPT-6 Astra Extra High	OpenAI	74.1%	2026-11-11
2	Gemini 3.8 Flash High	Google	73.8%	2026-11-11
3	Claude Opus 5 Max	Anthropic	73.7%	2026-11-11
4	GPT-6 Astra Max	OpenAI	73.2%	2026-11-11
5	GPT-6 Astra High	OpenAI	73.2%	2026-11-11
6	Claude Opus 5 Extra High	Anthropic	73.2%	2026-11-11
7	Claude Opus 5 High	Anthropic	72.8%	2026-11-11
8	GPT-6 Astra Medium	OpenAI	72.8%	2026-11-11
9	GPT-5.6 Sol Max	OpenAI	72.7%	2026-11-11
10	Gemini 3.8 Flash Medium	Google	71.0%	2026-11-11
11	GPT-5.6 Sol Extra High	OpenAI	70.7%	2026-11-11
12	Claude Fable 5 Extra High	Anthropic	69.9%	2026-11-11
13	Claude Fable 5 Max	Anthropic	69.7%	2026-11-11
14	GPT-5.6 Terra Max	OpenAI	69.6%	2026-11-11
15	GPT-5.6 Sol High	OpenAI	69.4%	2026-11-11
16	GLM-5.3 Max	Zhipu AI	69.0%	2026-11-11
17	Claude Opus 5 Medium	Anthropic	68.9%	2026-11-11
18	Claude Fable 5 High	Anthropic	68.6%	2026-11-11
19	Kimi K3 Max	Moonshot AI	68.5%	2026-11-11
20	Grok 4.6 Medium	xAI	67.5%	2026-11-11
21	GPT-5.6 Luna Max	OpenAI	67.2%	2026-11-11
22	GPT-5.5 Extra High	OpenAI	67.0%	2026-11-11
23	GPT-6 Astra Low	OpenAI	67.0%	2026-11-11
24	Grok 4.6 Extra High	xAI	66.7%	2026-11-11
25	Gemini 3.7 Flash Medium	Google	65.5%	2026-11-11
26	Claude Fable 5 Medium	Anthropic	65.4%	2026-11-11
27	Gemini 3.7 Flash High	Google	65.3%	2026-11-11
28	Grok 4.6 High	xAI	65.2%	2026-11-11
29	GPT-5.5 High	OpenAI	64.4%	2026-11-11
30	GLM-5.3-Flash Max	Zhipu AI	63.4%	2026-11-11
31	GPT-5.6 Sol Medium	OpenAI	61.1%	2026-11-11
32	GPT-5.6 Terra Extra High	OpenAI	60.2%	2026-11-11
33	Claude Fable 5 Low	Anthropic	59.6%	2026-11-11
34	Claude Opus 4.8 Max	Anthropic	59.0%	2026-11-11
35	Claude Opus 5 Low	Anthropic	58.1%	2026-11-11
36	Qwen3.8 Max Extra High	Alibaba	57.5%	2026-11-11
37	GPT-5.6 Luna Extra High	OpenAI	56.9%	2026-11-11
38	Muse Spark 1.2 Extra High	Meta	54.9%	2026-11-11
39	Claude Opus 4.8 Extra High	Anthropic	54.4%	2026-11-11
40	GPT-5.5 Medium	OpenAI	54.0%	2026-11-11
41	Claude Sonnet 5 Max	Anthropic	53.9%	2026-11-11
42	Grok 4.5 High	xAI	53.8%	2026-11-11
43	GPT-5.6 Terra High	OpenAI	53.8%	2026-11-11
44	Gemini 3.7 Flash Low	Google	53.8%	2026-11-11
45	Muse Spark 1.1 Extra High	Meta	53.3%	2026-11-11
46	GPT-5.4 Extra High	OpenAI	51.8%	2026-11-11
47	Claude Opus 4.8 High	Anthropic	51.8%	2026-11-11
48	Claude Sonnet 5 Extra High	Anthropic	49.7%	2026-11-11
49	Claude Opus 4.8 Medium	Anthropic	48.7%	2026-11-11
50	Claude Sonnet 5 High	Anthropic	48.2%	2026-11-11
51	Gemini 3.6 Flash High	Google	46.7%	2026-11-11
52	GPT-5.6 Sol Low	OpenAI	45.4%	2026-11-11
53	GPT-5.6 Luna High	OpenAI	44.3%	2026-11-11
54	GLM-5.2 Max	Zhipu AI	43.8%	2026-11-11
55	Grok 4.6 Low	xAI	41.7%	2026-11-11
56	Claude Opus 4.8 Low	Anthropic	40.8%	2026-11-11
57	Claude Sonnet 5 Medium	Anthropic	39.8%	2026-11-11
58	Gemini 3.5 Flash Medium	Google	37.4%	2026-11-11
59	GLM-5.2 High	Zhipu AI	36.3%	2026-11-11
60	Gemini 3.5 Flash High	Google	36.1%	2026-11-11
61	GPT-5.6 Terra Medium	OpenAI	35.1%	2026-11-11
62	Kimi K2.7 Code	Moonshot AI	30.5%	2026-11-11
63	Claude Sonnet 5 Low	Anthropic	30.5%	2026-11-11
64	Claude Sonnet 4.6 High	Anthropic	29.9%	2026-11-11
65	GPT-5.5 Low	OpenAI	27.0%	2026-11-11
66	GPT-5.6 Terra Low	OpenAI	24.1%	2026-11-11
67	Gemini 3.1 Pro Preview High	Google	11.7%	2026-11-11
68	GPT-5.6 Luna Medium	OpenAI	11.3%	2026-11-11
69	GPT-5.6 Luna Low	OpenAI	1.6%	2026-11-11