Your AI Benchmark Observatory

all the models

Independent signal across paid and open-weight models — GPQA, SWE-bench, Arena Elo, and more, kept current from live sources.

Explore models Compare

AI Race

Trailing twelve months, week by week. Each lab’s line is its highest-scoring closed GPQA model released by that week.

AnthropicGoogleOpenAIxAIMetaByteDanceAlibaba Cloud

1OpenAI0.960
GPT-6 Astra
2Anthropic0.946
Claude Mythos Preview
3Google0.943
Gemini 3.1 Pro
4xAI0.930
Grok 4.5
5Alibaba Cloud0.924
Qwen3.7 Max
6Meta0.895
Muse Spark
7ByteDance0.889
Seed 2.0 Pro

Weeks in the lead

Same trailing twelve months · 52 weeks

OpenAI21
Anthropic17
Google14
Alibaba Cloud0
ByteDance0
Meta0
xAI0

Week	Leader	Anthropic	Google	OpenAI	xAI	Meta	ByteDance	Alibaba Cloud
Nov 2	OpenAI	0.946	0.943	0.960	0.930	0.895	0.889	0.924
Oct 26	OpenAI	0.946	0.943	0.960	0.930	0.895	0.889	0.924
Oct 19	OpenAI	0.946	0.943	0.960	0.930	0.895	0.889	0.924
Oct 12	OpenAI	0.946	0.943	0.960	0.930	0.895	0.889	0.924
Oct 5	OpenAI	0.946	0.943	0.960	0.930	0.895	0.889	0.924
Sep 28	OpenAI	0.946	0.943	0.960	0.930	0.895	0.889	0.924
Sep 21	OpenAI	0.946	0.943	0.960	0.930	0.895	0.889	0.924
Sep 14	OpenAI	0.946	0.943	0.960	0.930	0.895	0.889	0.924
Sep 7	OpenAI	0.946	0.943	0.960	0.930	0.895	0.889	0.924
Aug 31	OpenAI	0.946	0.943	0.960	0.930	0.895	0.889	0.924
Aug 24	Anthropic	0.946	0.943	0.946	0.930	0.895	0.889	0.924
Aug 17	Anthropic	0.946	0.943	0.946	0.930	0.895	0.889	0.924
Aug 10	Anthropic	0.946	0.943	0.946	0.930	0.895	0.889	0.924
Aug 3	Anthropic	0.946	0.943	0.946	0.930	0.895	0.889	0.924
Jul 27	Anthropic	0.946	0.943	0.946	0.930	0.895	0.889	0.924
Jul 20	Anthropic	0.946	0.943	0.946	0.930	0.895	0.889	0.924
Jul 13	Anthropic	0.946	0.943	0.946	0.930	0.895	0.889	0.924
Jul 6	Anthropic	0.946	0.943	0.946	0.884	0.895	0.889	0.924
Jun 29	Anthropic	0.946	0.943	0.936	0.884	0.895	0.889	0.924
Jun 22	Anthropic	0.946	0.943	0.936	0.884	0.895	0.889	0.924
Jun 15	Anthropic	0.946	0.943	0.936	0.884	0.895	0.889	0.924
Jun 8	Anthropic	0.946	0.943	0.936	0.884	0.895	0.889	0.924
Jun 1	Anthropic	0.946	0.943	0.936	0.884	0.895	0.889	0.924
May 25	Anthropic	0.946	0.943	0.936	0.884	0.895	0.889	0.924
May 18	Anthropic	0.946	0.943	0.936	0.884	0.895	0.889	0.924
May 11	Anthropic	0.946	0.943	0.936	0.884	0.895	0.889	0.904
May 4	Anthropic	0.946	0.943	0.936	0.884	0.895	0.889	0.904
Apr 27	Google	0.942	0.943	0.936	0.875	0.895	0.889	0.904
Apr 20	Google	0.942	0.943	0.936	0.875	0.895	0.889	0.904
Apr 13	Google	0.942	0.943	0.932	0.875	0.895	0.889	0.904
Apr 6	Google	0.913	0.943	0.932	0.875	0.895	0.889	0.904
Mar 30	Google	0.913	0.943	0.932	0.875	—	0.889	0.904
Mar 23	Google	0.913	0.943	0.932	0.875	—	0.889	0.874
Mar 16	Google	0.913	0.943	0.932	0.875	—	0.889	0.874
Mar 9	Google	0.913	0.943	0.932	0.875	—	0.889	0.874
Mar 2	Google	0.913	0.943	0.932	0.875	—	0.889	0.874
Feb 23	Google	0.913	0.943	0.932	0.875	—	0.889	0.874
Feb 16	Google	0.913	0.943	0.932	0.875	—	0.889	0.874
Feb 9	OpenAI	0.913	0.919	0.932	0.875	—	0.889	0.874
Feb 2	OpenAI	0.913	0.919	0.932	0.875	—	—	0.620
Jan 26	OpenAI	0.870	0.919	0.932	0.875	—	—	0.620
Jan 19	OpenAI	0.870	0.919	0.932	0.875	—	—	0.620
Jan 12	OpenAI	0.870	0.919	0.932	0.875	—	—	0.620
Jan 5	OpenAI	0.870	0.919	0.932	0.875	—	—	0.620
Dec 29	OpenAI	0.870	0.919	0.932	0.875	—	—	0.620
Dec 22	OpenAI	0.870	0.919	0.932	0.875	—	—	0.620
Dec 15	OpenAI	0.870	0.919	0.932	0.875	—	—	0.620
Dec 8	OpenAI	0.870	0.919	0.932	0.875	—	—	—
Dec 1	Google	0.870	0.919	0.881	0.875	—	—	—
Nov 24	Google	0.870	0.919	0.881	0.875	—	—	—
Nov 17	Google	0.848	0.919	0.881	0.875	—	—	—
Nov 10	OpenAI	0.848	0.864	0.881	0.875	—	—	—

Live ranking plane

Sorted by GPQA when available. Multi-source columns stay transparent.

442 models · 772 benches · 2515 scores · synced 11/3/2026, 8:30:13 AM

#	Model	GPQA	SWE	MMLU-Pro	HLE	Arena	AA Idx	$/M in
1	GPT-6 AstraOpenAIclosed	0.960	—	—	—	—	—	10.00
2	Claude Mythos PreviewAnthropicclosed	0.946	0.939	—	0.647	—	—	—
3	GPT-5.6 SolOpenAIclosed	0.946	—	—	—	—	—	5.00
4	Gemini 3.1 ProGoogleclosed	0.943	0.806	—	0.514	1487	—	2.00
5	Claude Opus 4.7Anthropicclosed	0.942	0.876	—	0.547	1494	—	5.00
6	Kimi K3.1 ProMoonshot AIclosed	0.942	—	—	0.511	—	—	2.00
7	Claude Opus 4.8Anthropicclosed	0.936	0.886	—	0.579	1475	—	5.00
8	GPT-5.5OpenAIclosed	0.936	—	—	0.522	1477	—	5.00
9	Kimi K3Moonshot AIopen	0.935	—	—	0.560	—	—	2.85
10	GPT-5.2 ProOpenAIclosed	0.932	—	—	0.366	—	—	—
11	Grok 4.5xAIclosed	0.930	—	—	—	1466	—	2.00
12	GPT-5.6 TerraOpenAIclosed	0.929	—	—	—	—	—	2.00
13	GPT-5.4OpenAIclosed	0.928	—	—	0.398	1466	—	2.50
14	Qwen3.8 MaxAlibaba Cloud / Qwen Teamopen	0.926	—	—	0.436	1482	—	1.65
15	GPT-5.2OpenAIclosed	0.924	0.800	—	0.345	—	—	1.75
16	Qwen3.7 MaxAlibaba Cloud / Qwen Teamclosed	0.924	0.804	0.896	0.414	—	—	1.25
17	GPT-5.6 LunaOpenAIclosed	0.923	—	—	—	—	—	0.20
18	Hy4 previewTencentopen	0.923	—	—	—	—	—	—
19	Gemini 3 ProGoogleclosed	0.919	0.762	—	0.458	1485	—	—
20	Qwen3.8 FlashAlibaba Cloud / Qwen Teamclosed	0.917	—	—	0.359	—	—	0.15
21	Qwen3.8-Flash-NextAlibaba Cloud / Qwen Teamopen	0.917	—	—	0.359	—	—	—
22	Claude Opus 4.6Anthropicclosed	0.913	0.808	—	0.531	1497	—	5.00
23	GLM-5.2Zhipu AIopen	0.912	—	—	0.547	—	—	0.75
24	DeepSeek-V4.1-FlashDeepSeekopen	0.909	—	—	0.368	—	—	0.22
25	Beam 1.0Reflection AIopen	0.906	0.816	—	0.360	—	—	0.40
26	Kimi K2.6Moonshot AIopen	0.905	0.802	—	0.364	—	—	0.75
27	Gemini 3 FlashGoogleclosed	0.904	0.780	—	0.435	1473	—	0.50
28	Hy3Tencentopen	0.904	0.780	—	—	—	—	0.14
29	Particle-1Particle AIclosed	0.904	—	—	—	—	—	0.25
30	Qwen3.6 PlusAlibaba Cloud / Qwen Teamclosed	0.904	0.788	0.885	0.288	—	—	0.50
31	Qwen3.7-PlusAlibaba Cloud / Qwen Teamclosed	0.903	0.777	0.885	0.347	—	—	—
32	DeepSeek-V4-Pro-MaxDeepSeekopen	0.901	0.806	0.875	0.482	—	—	1.30
33	Claude Sonnet 4.6Anthropicclosed	0.899	0.796	—	0.490	1472	—	3.00
34	Nemotron 3.5 Super VL (120B A12B)NVIDIAopen	0.895	0.711	—	0.318	—	—	0.20
35	Inkling-SmallThinking Machines Labopen	0.895	0.802	—	0.316	—	—	0.30
36	Muse SparkMetaclosed	0.895	0.774	—	0.584	1489	—	—
37	Mercury 3Inceptionclosed	0.892	—	—	—	—	—	0.10
38	Qwen3.8-27BAlibaba Cloud / Qwen Teamopen	0.892	—	—	0.308	—	—	0.40
39	Muse Spark OpenMetaopen	0.890	—	—	—	—	—	1.25
40	Solar Pro 4Upstageclosed	0.890	0.706	0.863	—	—	—	0.30