VECTOR WIREAI INTELLIGENCE
NVDA$1,847+3.2%MSFT$512+1.1%GOOGL$199-0.4%META$728+2.7%AMD$184-1.2%TSM$212+0.6%PLTR$98+4.1%AI IDX4,821+1.9%
PKT
SEEDRefresh Models Deals Regulatory Sources

naturalquestions_closedbook

63 models tracked

Natural Questions — HELM Lite explicit closed-book variant; kept apart from unscoped entry.

#ModelVendorBest scoreRunsLast seen
1Claude 3.5 SonnetAnthropic50.212026-05-10
2GPT-4oOpenAI50.122026-05-10
3GPT-4 TurboOpenAI48.222026-06-18
4Llama 3 (70B)Meta47.512026-05-10
5DeepSeek-V3DeepSeek46.712026-05-10
6GPT4OpenAI45.722026-05-10
7Llama 3.2 Vision Instruct Turbo (90B)Meta45.712026-05-10
8Palmyra-X-004Writer45.712026-05-10
9Llama 3.1 Instruct 405BMeta45.612026-05-10
10Gemini 1.5 ProGoogle45.522026-05-10
11Mistral Large 2Mistral45.312026-05-10
12Llama 3.1 Instruct Turbo (70B)Meta45.212026-05-10
13Gemini 2.0 Flash (experimental)Google44.312026-06-24
14Gemini 2.0 Flash ExpGoogle44.312026-05-10
15Claude 3 OpusAnthropic4412026-05-10
16Llama (65B)Meta43.312026-05-10
17Llama 3.3 Instruct Turbo (70B)Meta43.112026-05-10
18Palmyra X V2 (33B)Writer42.812026-05-10
19Yi Large (Preview)01.AI42.812026-05-10
20DeepSeek-ChatDeepSeek41.212026-05-10
21Palmyra X V3 (72B)Writer40.712026-05-10
22Nova Pro (Non-Reasoning)Amazon40.522026-06-19
23Jamba 1.5 LargeAI2139.412026-05-10
24CommandCohere39.112026-05-10
25Arctic InstructSnowflake3912026-05-10
26Qwen2 Instruct (72B)Alibaba3912026-05-10
27Jamba 1.5 MiniAI2138.812026-05-10
28GPT-4o miniOpenAI38.512026-05-10
29Jurassic 2 Jumbo (178B)AI2138.512026-05-10
30Jamba InstructAI2138.412026-05-10
31Llama 3 Instruct 8BMeta37.812026-05-10
32Mistral V0.1 (7B)Mistral36.712026-05-10
33Qwen2.5 Instruct 72BAlibaba35.912026-05-10
34Gemma 2 Instruct (27B)Google35.312026-05-10
35Amazon Nova LiteAmazon35.212026-06-23
36Command RCohere35.212026-05-10
37Jurassic 2 Grande (17B)AI213512026-05-10
38Claude 3.5 HaikuAnthropic34.422026-06-19
39Command R+ (Apr '24)Cohere34.312026-05-10
40Gemma (7B)Google33.612026-05-10
41Gemini 1.5 Flash (001)Google33.222026-05-10
42Gemma 2 Instruct (9B)Google32.812026-05-10
43Phi 3 Small 8K InstructMicrosoft32.412026-05-10
44Mistral LargeMistral31.112026-05-10
45Mistral Small (2402)Mistral30.412026-05-10
46Luminous Supreme (70B)Aleph Alpha29.912026-05-10
47Solar ProUpstage29.712026-05-10
48Mistral Medium (Non-Reasoning)Mistral2912026-06-23
49Nova Micro (Non-Reasoning)Amazon28.522026-06-19
50DBRX InstructDatabricks28.412026-05-10
51Phi 3 Medium 4K InstructMicrosoft27.812026-05-10
52Mistral NemoMistral26.512026-05-10
53OLMo (7B)AllenAI25.912026-05-10
54Mistral Instruct V0.3 (7B)Mistral25.312026-05-10
55Luminous Extended (30B)Aleph Alpha25.312026-05-10
56Llama 3.2 Vision Instruct Turbo (11B)Meta23.412026-05-10
57Llama 3.1 Instruct Turbo (8B)Meta20.912026-05-10
58Qwen2.5 Instruct Turbo (7B)Alibaba20.512026-05-10
59Luminous Base (13B)Aleph Alpha19.712026-05-10
60Command LightCohere19.512026-05-10
61Phi 2Microsoft15.512026-05-10
62Claude 3 HaikuAnthropic14.412026-05-10
63Claude 3 SonnetAnthropic2.812026-05-10

Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.