Agentic · tools

GDP.pdf leaderboard

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

As of 2026-10-07, the highest measured score on GDP.pdf is 31.0% by GPT-6.1 Sol. 133 more models have estimated scores, calibrated from the benchmarks they were measured on.

Measured scores: benchlm.ai.

#	Model	Score	Source
1	GPT-6.1 Sol	31.0%	measured
2	GPT-6 Astra	31.0%	measured
3	Claude Opus 5	27.1%	estimated ± 3.8 pp, high confidence
4	Muse Spark 1.3	26.6%	measured
5	Claude Fable 5.1	26.2%	measured
6	Claude Opus 5.5	26.2%	measured
7	Claude Sonnet 5.5	25.8%	measured
8	GPT-5.5	25.5%	estimated ± 3.8 pp, high confidence
9	GPT-6 Sol	25.2%	measured
10	Qwen3.8 Max	24.9%	estimated ± 3.4 pp, high confidence
11	Qwen3.8-Flash-Next	24.4%	estimated ± 3.4 pp, high confidence
12	GPT-5.6 Sol	24.4%	estimated ± 3.8 pp, high confidence
13	Claude Sonnet 5	22.9%	estimated ± 4.4 pp, low confidence
14	GPT-5.6 Terra	22.9%	estimated ± 4.4 pp, low confidence
15	GPT-6 Luna	22.8%	measured
16	Ling 3.1 Flash	22.5%	estimated ± 5.4 pp, medium confidence
17	Qwen3.8 Max Preview	22.5%	estimated ± 5.4 pp, medium confidence
18	Muse Spark 1.2	22.5%	estimated ± 5.4 pp, medium confidence
19	GPT-5.4	22.4%	estimated ± 4.4 pp, low confidence
20	Claude Opus 4.7	22.0%	estimated ± 4.4 pp, low confidence
21	GPT-5.6 Luna	22.0%	estimated ± 4.4 pp, low confidence
22	Kimi K3	22.0%	measured
23	Gemini 4 Argon	21.8%	measured
24	Claude Opus 4.6	21.8%	estimated ± 4.4 pp, low confidence
25	Claude Sonnet 4.6	21.5%	estimated ± 4.4 pp, low confidence
26	Gemini 3.8 Flash	21.0%	measured
27	Claude Haiku 5.5	20.8%	measured
28	Claude Opus 4.8	20.6%	estimated ± 4.9 pp, high confidence
29	Grok 4.7	20.0%	measured
30	Claude Opus 4.7 (Adaptive)	19.8%	estimated ± 4.9 pp, high confidence
31	Claude Mythos Preview	19.3%	estimated ± 5.9 pp, low confidence
32	MiMo-V2.6-Pro	19.2%	measured
33	Mistral Large 4	18.6%	measured
34	GPT-5.5 Pro	18.5%	estimated ± 6.6 pp, low confidence
35	Muse Spark 1.1	18.1%	estimated ± 4.9 pp, high confidence
36	Atria Dawn Preview	17.8%	estimated ± 5.2 pp, low confidence
37	Claude 4.1 Opus	17.8%	estimated ± 5.2 pp, low confidence
38	Claude 4 Sonnet	17.8%	estimated ± 5.2 pp, low confidence
39	Claude Haiku 4.5	17.8%	estimated ± 5.2 pp, low confidence
40	Claude Opus 4.5	17.8%	estimated ± 5.2 pp, low confidence
41	Claude Sonnet 4.5	17.8%	estimated ± 5.2 pp, low confidence
42	Gemini 3 Flash	17.8%	estimated ± 5.2 pp, low confidence
43	Gemini 3 Pro	17.8%	estimated ± 5.2 pp, low confidence
44	GPT-5.1-Codex	17.8%	estimated ± 5.2 pp, low confidence
45	GPT-5.2	17.8%	estimated ± 5.2 pp, low confidence
46	GPT-5.2-Codex	17.8%	estimated ± 5.2 pp, low confidence
47	GPT-5.3 Codex	17.8%	estimated ± 5.2 pp, low confidence
48	GPT-5 (high)	17.8%	estimated ± 5.2 pp, low confidence
49	Kimi K2.5	17.8%	estimated ± 5.2 pp, low confidence
50	Qwen3.5 Plus	17.8%	estimated ± 5.2 pp, low confidence
51	Claude Fable 5	17.6%	estimated ± 3.7 pp, low confidence
52	Gemini 3.5 Flash	17.3%	estimated ± 3.7 pp, low confidence
53	Grok 4.6	16.7%	estimated ± 3.7 pp, low confidence
54	Qwen3.8-27B	16.6%	measured
55	GLM-5.3-Flash	15.4%	measured
56	GPT-5.4 Pro	15.3%	estimated ± 6.6 pp, low confidence
57	Step 5 Preview	14.8%	measured
58	MiMo-V2.6-Flash	13.9%	estimated ± 3.4 pp, high confidence
59	Gemini 3.7 Flash	13.3%	estimated ± 3.4 pp, high confidence
60	DeepSeek V4 Pro 0813	13.0%	estimated ± 3.4 pp, medium confidence
61	DeepSeek V4.1 Flash	12.8%	measured
62	Inkling	12.8%	measured
63	DeepSeek V4 Flash 0731	12.8%	estimated ± 3.4 pp, medium confidence
64	Claude Mythos 5	12.2%	estimated ± 6.6 pp, low confidence
65	Hy4 preview	11.6%	estimated ± 3.4 pp, medium confidence
66	GLM-5.3	11.2%	measured
67	Ornith-1.5-397B	10.6%	estimated ± 6.6 pp, low confidence
68	Grok 4.5	10.2%	estimated ± 5.4 pp, medium confidence
69	Kimi K2.6	10.1%	estimated ± 4.9 pp, high confidence
70	Muse Glimmer 30B	10.0%	measured
71	MiniMax M3	9.8%	measured
72	GLM-5.2	9.6%	estimated ± 5.4 pp, medium confidence
73	dots3-note Preview	9.5%	estimated ± 6.6 pp, low confidence
74	A.X K2	9.4%	estimated ± 5.4 pp, medium confidence
75	Apodex 1.1	9.4%	estimated ± 5.4 pp, medium confidence
76	Apodex 1.1 Mini	9.4%	estimated ± 5.4 pp, medium confidence
77	Celeris-1	9.4%	estimated ± 5.4 pp, low confidence
78	Command A+	9.4%	estimated ± 5.4 pp, low confidence
79	DeepSeek V3	9.4%	estimated ± 5.4 pp, low confidence
80	DeepSeek V3 0324	9.4%	estimated ± 5.4 pp, low confidence
81	Gemini 2.5 Pro	9.4%	estimated ± 5.4 pp, low confidence
82	Gemini 3.1 Pro	9.4%	estimated ± 5.4 pp, medium confidence
83	Gemini 3.5 Flash-Lite	9.4%	estimated ± 5.4 pp, medium confidence
84	Gemini 3.6 Flash	9.4%	estimated ± 5.4 pp, medium confidence
85	Gemma 3 27B	9.4%	estimated ± 5.4 pp, low confidence
86	Gemma 4 12B	9.4%	estimated ± 5.4 pp, low confidence
87	Gemma 4 26B A4B	9.4%	estimated ± 5.4 pp, low confidence
88	Gemma 4 31B	9.4%	estimated ± 5.4 pp, low confidence
89	Gemma 4 E2B	9.4%	estimated ± 5.4 pp, low confidence
90	Gemma 4 E4B	9.4%	estimated ± 5.4 pp, low confidence
91	GLM-4.7	9.4%	estimated ± 5.4 pp, medium confidence
92	GLM-5.1	9.4%	estimated ± 5.4 pp, medium confidence
93	GPT-4.1 mini	9.4%	estimated ± 5.4 pp, low confidence
94	GPT-4.1 nano	9.4%	estimated ± 5.4 pp, low confidence
95	GPT-4o	9.4%	estimated ± 5.4 pp, low confidence
96	GPT-4o mini	9.4%	estimated ± 5.4 pp, low confidence
97	GPT-5.1	9.4%	estimated ± 5.4 pp, medium confidence
98	GPT-5.4 mini	9.4%	estimated ± 5.4 pp, medium confidence
99	GPT-5.4 nano	9.4%	estimated ± 5.4 pp, medium confidence
100	GPT-OSS 120B	9.4%	estimated ± 5.4 pp, low confidence
101	GPT-OSS 20B	9.4%	estimated ± 5.4 pp, low confidence
102	Granite 4.2 30B	9.4%	estimated ± 5.4 pp, low confidence
103	Granite 4.2 3B	9.4%	estimated ± 5.4 pp, low confidence
104	Granite 4.2 8B	9.4%	estimated ± 5.4 pp, low confidence
105	Grok 4.3	9.4%	estimated ± 5.4 pp, medium confidence
106	Hy3	9.4%	estimated ± 5.4 pp, medium confidence
107	Hy3 Preview	9.4%	estimated ± 5.4 pp, medium confidence
108	Inkling-Small	9.4%	estimated ± 5.4 pp, medium confidence
109	K-Exaone	9.4%	estimated ± 5.4 pp, low confidence
110	Kimi K2.5 (Reasoning)	9.4%	estimated ± 5.4 pp, medium confidence
111	Kimi K2.7 Code	9.4%	estimated ± 5.4 pp, medium confidence
112	LFM2.5-2.6B	9.4%	estimated ± 5.4 pp, low confidence
113	Ling 2.6 Flash	9.4%	estimated ± 5.4 pp, low confidence
114	Ling 3.0 Flash	9.4%	estimated ± 5.4 pp, medium confidence
115	Ling 3.0 Flash FP8	9.4%	estimated ± 5.4 pp, medium confidence
116	Ling 3.0 Flash VL	9.4%	estimated ± 5.4 pp, medium confidence
117	Ling 3.0 Tiny	9.4%	estimated ± 5.4 pp, low confidence
118	Llama 4 Maverick	9.4%	estimated ± 5.4 pp, low confidence
119	Llama 4 Scout	9.4%	estimated ± 5.4 pp, low confidence
120	Mercury 2.5	9.4%	estimated ± 5.4 pp, low confidence
121	MiMo-V2.5-Pro	9.4%	estimated ± 5.4 pp, medium confidence
122	MiMo-V2-Flash	9.4%	estimated ± 5.4 pp, low confidence
123	MiniCPM5-2B	9.4%	estimated ± 5.4 pp, low confidence
124	MiniMax M2.7	9.4%	estimated ± 5.4 pp, medium confidence
125	Mistral Large 3	9.4%	estimated ± 5.4 pp, low confidence
126	Mistral Medium 3.5 128B	9.4%	estimated ± 5.4 pp, low confidence
127	Mistral Small 4	9.4%	estimated ± 5.4 pp, low confidence
128	Mistral Small 4 (Reasoning)	9.4%	estimated ± 5.4 pp, low confidence
129	Muse Spark	9.4%	estimated ± 5.4 pp, medium confidence
130	Nemotron 3.5 Lightning 30B A3B NVFP4	9.4%	estimated ± 5.4 pp, low confidence
131	Nemotron 3 Nano 30B	9.4%	estimated ± 5.4 pp, low confidence
132	Nemotron 3 Nano Omni 30B A3B	9.4%	estimated ± 5.4 pp, low confidence
133	Nemotron 3 Super 100B	9.4%	estimated ± 5.4 pp, low confidence
134	North Mini Code	9.4%	estimated ± 5.4 pp, low confidence
135	Quasar 438B	9.4%	estimated ± 5.4 pp, medium confidence
136	Qwen3.5-122B-A10B	9.4%	estimated ± 5.4 pp, medium confidence
137	Qwen3.6-27B	9.4%	estimated ± 5.4 pp, medium confidence
138	Qwen3.6-35B-A3B	9.4%	estimated ± 5.4 pp, medium confidence
139	Qwen3.6 Plus	9.4%	estimated ± 5.4 pp, medium confidence
140	Qwen3.7 Max	9.4%	estimated ± 5.4 pp, medium confidence
141	Solar Pro 3	9.4%	estimated ± 5.4 pp, low confidence
142	Solar Pro 4	9.4%	estimated ± 5.4 pp, medium confidence
143	Step 3.7 Flash	9.4%	estimated ± 5.4 pp, medium confidence
144	Trinity-Large-Preview	9.4%	estimated ± 5.4 pp, low confidence
145	Trinity-Large-Thinking	9.4%	estimated ± 5.4 pp, low confidence
146	Ultravox v0.6 Llama 3.3 70B	9.4%	estimated ± 5.4 pp, low confidence
147	Beam	9.3%	estimated ± 6.6 pp, low confidence
148	Agents-A1	9.3%	estimated ± 6.6 pp, low confidence
149	Agents-A1-4B	9.3%	estimated ± 6.6 pp, low confidence
150	Ornith-1.5-35B-A3B	9.3%	estimated ± 6.6 pp, low confidence
151	LongCat-Flash-Lite-Sparse	9.3%	estimated ± 6.6 pp, low confidence
152	Ornith-1.5-9B	9.3%	estimated ± 6.6 pp, low confidence
153	Qwen3.5-27B	9.3%	estimated ± 6.6 pp, low confidence
154	Qwen3.5-35B-A3B	9.3%	estimated ± 6.6 pp, low confidence
155	Qwen3.5 397B	9.3%	estimated ± 6.6 pp, low confidence
156	Qwen3.7 Plus	7.1%	estimated ± 4.9 pp, medium confidence
157	Nemotron 3 Ultra	5.0%	measured

GDP.pdf leaderboard

All leaderboards

Agentic · terminal

Agentic · tools

Coding

Math

Knowledge & reasoning

Instruction following

Multilingual

Vision & documents

Long context

General