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

Terminal-Bench

27 models tracked

Terminal-Bench — variant "agent:terminus-1-framework" confirms v1 (Terminus-1 harness), distinct from 2.0/2.1

#ModelVendorBest scoreRunsLast seen
1Claude Sonnet 4.5Anthropic5132026-08-23
2MiniMax M2.1MiniMax47.912026-08-23
3Kimi K2 (Reasoning)Moonshot47.142026-06-06
4MiniMax M2MiniMax46.322026-08-23
5GPT-5OpenAI43.822026-06-06
6Claude Opus 4.1Anthropic43.312026-08-23
7Claude 4 OpusAnthropic43.252026-08-23
8Claude Haiku 4.5Anthropic4112026-08-23
9GLM-4.6Z.ai40.522026-08-23
10GLM-4.5Z.ai39.922026-08-23
11DeepSeek-V3.2DeepSeek37.722026-06-06
12DeepSeek-V3.2-ExpDeepSeek37.712026-08-23
13Qwen3 Coder 480B A35B InstructAlibaba37.512026-05-15
14K2-Instruct-0711Moonshot37.512026-05-15
15Claude Sonnet 4Anthropic36.462026-08-23
16Claude 3.7 SonnetAnthropic35.212026-08-23
17LongCat Flash LiteMeituan33.812026-08-23
18GLM-4.7Z.ai33.312026-08-23
19DeepSeek-V3.1DeepSeek31.322026-08-23
20MiMo V2 FlashXiaomi30.512026-08-23
21GPT-4.1OpenAI30.322026-06-15
22Kimi K2 InstructMoonshot3042026-08-23
23GLM-4.5-AirZ.ai3012026-08-23
24Nemotron 3 Super 120B A12BNVIDIA25.812026-08-23
25Gemini 2.5 ProGoogle25.312026-06-06
26NVIDIA Nemotron 3 Nano 30B A3BNVIDIA8.512026-08-23
27DeepSeek-R1DeepSeek5.712026-08-23

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