Coding

FrontierCode 1.1 Extended leaderboard

Name: benchgap: LLM benchmark scores, measured and estimated
Creator: benchgap

As of 2026-10-07, the highest measured score on FrontierCode 1.1 Extended is 64.5% by GPT-6 Astra. 139 more models have estimated scores, calibrated from the benchmarks they were measured on.

Measured scores: benchlm.ai.

#	Model	Score	Source
1	GPT-6 Astra	64.5%	measured
2	Claude Opus 5	63.6%	measured
3	Claude Opus 5.5	63.6%	measured
4	Claude Fable 5.1	62.5%	estimated ± 1.6 pp, low confidence
5	Claude Fable 5	62.4%	estimated ± 1.6 pp, medium confidence
6	Gemini 3.8 Flash	62.3%	estimated ± 1.6 pp, medium confidence
7	Claude Mythos 5	61.5%	estimated ± 3.0 pp, medium confidence
8	Grok 4.6	61.3%	measured
9	DeepSeek V4 Pro 0813	61.1%	estimated ± 2.4 pp, medium confidence
10	Grok 4.7	61.0%	estimated ± 3.0 pp, medium confidence
11	GPT-5.6 Sol	60.6%	measured
12	Sakana Fugu-Ultra	59.9%	estimated ± 3.0 pp, medium confidence
13	Muse Spark 1.3	59.5%	estimated ± 3.0 pp, medium confidence
14	Grok 4.5	59.4%	estimated ± 1.6 pp, medium confidence
15	GLM-5.3	59.2%	estimated ± 2.4 pp, medium confidence
16	Claude Sonnet 5.5	59.1%	measured
17	Hy4 preview	57.8%	estimated ± 3.0 pp, medium confidence
18	Beam	57.7%	estimated ± 3.0 pp, medium confidence
19	Qwen3.8 Max Preview	57.7%	estimated ± 3.4 pp, low confidence
20	Ornith-1.5-397B	57.6%	estimated ± 3.0 pp, medium confidence
21	Claude Opus 4.7 (Adaptive)	57.4%	estimated ± 3.0 pp, medium confidence
22	Qwen3.8-Omni-Flash	57.1%	estimated ± 3.0 pp, medium confidence
23	Qwen3.8-Flash-Next	56.8%	estimated ± 3.0 pp, low confidence
24	Ornith-1.0-397B	56.7%	estimated ± 3.0 pp, low confidence
25	dots3-note Preview	56.3%	estimated ± 3.0 pp, low confidence
26	Atria Dawn Preview	55.9%	estimated ± 3.0 pp, low confidence
27	Ornith-1.5-35B-A3B	55.9%	estimated ± 3.0 pp, low confidence
28	Laguna S 2.1	55.8%	estimated ± 3.0 pp, low confidence
29	GPT-5.6 Terra	55.8%	measured
30	Sakana Fugu	55.7%	estimated ± 3.0 pp, low confidence
31	GPT-5.4	55.2%	estimated ± 3.0 pp, low confidence
32	Qwen3.7 Plus	55.2%	estimated ± 3.0 pp, low confidence
33	Claude Opus 4.8	55.2%	estimated ± 1.6 pp, medium confidence
34	Claude Sonnet 5	55.1%	estimated ± 1.6 pp, medium confidence
35	Kimi K3	55.1%	estimated ± 1.6 pp, low confidence
36	GPT-5.5	55.1%	estimated ± 1.6 pp, low confidence
37	Composer 2.5	55.1%	estimated ± 1.6 pp, low confidence
38	Gemini 3.5 Flash	55.1%	estimated ± 1.6 pp, low confidence
39	Gemini 3.6 Flash	55.1%	estimated ± 1.6 pp, low confidence
40	GLM-5.2	55.1%	estimated ± 1.6 pp, low confidence
41	Kimi K2.7 Code	55.1%	estimated ± 1.6 pp, low confidence
42	GPT-5.6 Luna	55.1%	measured
43	Claude Opus 4.5	55.0%	estimated ± 3.0 pp, low confidence
44	Step 3.7 Flash	54.7%	estimated ± 3.0 pp, low confidence
45	GPT-5.2	54.5%	estimated ± 3.0 pp, low confidence
46	GLM-5	54.3%	estimated ± 3.0 pp, low confidence
47	Claude Opus 4.6	53.7%	estimated ± 3.0 pp, low confidence
48	MAI-Thinking-1	53.4%	estimated ± 3.0 pp, low confidence
49	Muse Glimmer 30B	52.8%	estimated ± 3.0 pp, low confidence
50	GLM-5.3-Flash	52.7%	estimated ± 2.4 pp, low confidence
51	Qwen3.5 397B	52.7%	estimated ± 3.0 pp, low confidence
52	Kimi K2.5	52.6%	estimated ± 3.0 pp, low confidence
53	Ornith-1.0-35B	52.5%	estimated ± 3.0 pp, low confidence
54	Qwen3.6-35B-A3B	52.1%	estimated ± 3.0 pp, low confidence
55	Quasar 438B	51.4%	estimated ± 3.4 pp, low confidence
56	Laguna XS 2.1	51.3%	estimated ± 3.0 pp, low confidence
57	Ornith-1.5-9B	51.2%	estimated ± 3.0 pp, low confidence
58	Apodex 1.1	51.1%	estimated ± 3.4 pp, low confidence
59	Apodex 1.1 Mini	51.1%	estimated ± 3.4 pp, low confidence
60	Hy3	49.9%	estimated ± 3.4 pp, low confidence
61	Hy3 Preview	49.9%	estimated ± 3.4 pp, low confidence
62	Ornith-1.0-9B	49.1%	estimated ± 3.0 pp, low confidence
63	LongCat-Flash-Lite-Sparse	47.9%	estimated ± 3.0 pp, low confidence
64	DeepSeek V4 Flash 0731	46.5%	estimated ± 2.4 pp, low confidence
65	Ling 3.0 Flash FP8	44.5%	estimated ± 3.4 pp, low confidence
66	MiMo-V2-Flash	43.9%	estimated ± 3.4 pp, low confidence
67	Granite 4.2 30B	43.6%	estimated ± 3.0 pp, low confidence
68	GPT-5.1	43.6%	estimated ± 3.4 pp, low confidence
69	Muse Spark 1.2	42.3%	estimated ± 2.4 pp, low confidence
70	Kimi K2.5 (Reasoning)	41.8%	estimated ± 3.4 pp, low confidence
71	LLaDA2.2-flash	41.5%	estimated ± 3.0 pp, low confidence
72	Qwen3.8-27B	41.2%	estimated ± 2.4 pp, low confidence
73	Qwen3.5-122B-A10B	41.0%	estimated ± 3.4 pp, low confidence
74	Qwen3.8 Max	40.4%	estimated ± 2.4 pp, low confidence
75	Gemma 4 31B	39.4%	estimated ± 3.4 pp, low confidence
76	o1	36.6%	estimated ± 3.4 pp, low confidence
77	Gemma 4 26B A4B	36.3%	estimated ± 3.4 pp, low confidence
78	GPT-5 (high)	35.2%	estimated ± 3.4 pp, low confidence
79	Nemotron 3 Super 100B	35.1%	estimated ± 3.4 pp, low confidence
80	Inkling-Small	34.2%	estimated ± 2.4 pp, low confidence
81	Claude Opus 4.7	33.8%	estimated ± 2.4 pp, low confidence
82	Muse Spark 1.1	33.8%	estimated ± 2.4 pp, low confidence
83	o1-preview	32.3%	estimated ± 3.4 pp, low confidence
84	Granite 4.2 8B	31.9%	estimated ± 3.0 pp, low confidence
85	Gemini 3.7 Flash	31.7%	estimated ± 2.4 pp, low confidence
86	K-Exaone	30.7%	estimated ± 3.4 pp, low confidence
87	Gemma 4 12B	29.8%	estimated ± 3.4 pp, low confidence
88	GPT-OSS 120B	29.3%	estimated ± 3.4 pp, low confidence
89	Gemini 3.1 Pro	28.4%	estimated ± 2.4 pp, low confidence
90	Command A+	27.2%	estimated ± 3.4 pp, low confidence
91	GPT-5.3 Codex	27.1%	estimated ± 2.4 pp, low confidence
92	MiniCPM5-2B	26.4%	estimated ± 3.0 pp, low confidence
93	Inkling	26.4%	estimated ± 2.4 pp, low confidence
94	Nemotron 3.5 Lightning 30B A3B NVFP4	26.3%	estimated ± 3.4 pp, low confidence
95	Mistral Small 4	26.2%	estimated ± 3.4 pp, low confidence
96	Mistral Small 4 (Reasoning)	26.2%	estimated ± 3.4 pp, low confidence
97	Claude Sonnet 4.6	26.1%	estimated ± 2.4 pp, low confidence
98	Trinity-Large-Preview	25.4%	estimated ± 3.4 pp, low confidence
99	Trinity-Large-Thinking	25.4%	estimated ± 3.4 pp, low confidence
100	Ling 2.6 Flash	25.0%	estimated ± 3.4 pp, low confidence
101	GLM-5.1	24.5%	estimated ± 2.4 pp, low confidence
102	Kimi K2.6	24.2%	estimated ± 2.4 pp, low confidence
103	Gemini 1.5 Pro	23.5%	estimated ± 3.4 pp, low confidence
104	DeepSeek V3	23.0%	estimated ± 3.4 pp, low confidence
105	Gemini 3.5 Flash-Lite	22.4%	estimated ± 2.4 pp, low confidence
106	Gemini 3 Flash	22.4%	estimated ± 2.4 pp, low confidence
107	MiniMax M3	22.4%	estimated ± 2.4 pp, low confidence
108	GPT-4 Turbo	21.6%	estimated ± 3.4 pp, low confidence
109	Muse Spark	21.6%	estimated ± 2.4 pp, low confidence
110	MiMo-V2.5-Pro	21.0%	estimated ± 2.4 pp, low confidence
111	GPT-OSS 20B	20.9%	estimated ± 3.4 pp, low confidence
112	MiniMax M2.7	20.7%	estimated ± 2.4 pp, low confidence
113	GPT-4.1 mini	20.5%	estimated ± 3.4 pp, low confidence
114	Mistral Large 3	20.4%	estimated ± 3.4 pp, low confidence
115	Qwen3.6 Plus	20.2%	estimated ± 2.4 pp, low confidence
116	Claude 3 Opus	19.9%	estimated ± 3.4 pp, low confidence
117	GPT-5.4 mini	19.6%	estimated ± 2.4 pp, low confidence
118	Qwen 3.6 Max (preview)	19.4%	estimated ± 2.4 pp, low confidence
119	GPT-5.2-Codex	18.8%	estimated ± 2.4 pp, low confidence
120	Grok 4.20	18.6%	estimated ± 2.4 pp, low confidence
121	Grok 4.3	17.5%	estimated ± 2.4 pp, low confidence
122	MiMo-V2.5	17.0%	estimated ± 2.4 pp, low confidence
123	Llama 4 Maverick	16.8%	estimated ± 3.4 pp, low confidence
124	Qwen3.6-27B	15.8%	estimated ± 2.4 pp, low confidence
125	GPT-5.4 nano	15.6%	estimated ± 2.4 pp, low confidence
126	GLM-4.7	15.1%	estimated ± 2.4 pp, low confidence
127	Celeris-1	15.0%	estimated ± 3.4 pp, low confidence
128	Nemotron 3 Nano 30B	15.0%	estimated ± 3.4 pp, low confidence
129	Nemotron 3 Ultra	14.7%	estimated ± 2.4 pp, low confidence
130	Qwen3.7 Max	14.5%	estimated ± 2.4 pp, low confidence
131	Nemotron 3 Nano Omni 30B A3B	14.4%	estimated ± 3.4 pp, low confidence
132	Ultravox v0.6 Llama 3.3 70B	12.6%	estimated ± 3.4 pp, low confidence
133	Claude Haiku 4.5	12.2%	estimated ± 2.4 pp, low confidence
134	GPT-4o mini	12.1%	estimated ± 3.4 pp, low confidence
135	Mistral Medium 3.5 128B	12.0%	estimated ± 2.4 pp, low confidence
136	GPT-4.1 nano	11.8%	estimated ± 3.4 pp, low confidence
137	Ling 3.0 Flash	10.8%	estimated ± 2.4 pp, low confidence
138	Gemma 3 27B	10.8%	estimated ± 3.4 pp, low confidence
139	Qwen3.5 Flash	10.1%	estimated ± 2.4 pp, low confidence
140	Gemma 4 E4B	10.1%	estimated ± 3.4 pp, low confidence
141	Llama 4 Scout	8.8%	estimated ± 3.4 pp, low confidence
142	Gemini 3.1 Flash-Lite	8.8%	estimated ± 2.4 pp, low confidence
143	LFM2.5-2.6B	8.4%	estimated ± 3.4 pp, low confidence
144	Gemma 4 E2B	7.8%	estimated ± 3.4 pp, low confidence
145	Laguna M.1	5.4%	estimated ± 2.4 pp, low confidence
146	Laguna XS.2	4.2%	estimated ± 2.4 pp, low confidence
147	Gemini 2.5 Pro	3.9%	estimated ± 2.4 pp, low confidence

FrontierCode 1.1 Extended leaderboard

All leaderboards

Agentic · terminal

Agentic · tools

Coding

Math

Knowledge & reasoning

Instruction following

Multilingual

Vision & documents

Long context

General