DeepSeek 4.1
An open-source large language model released by DeepSeek, representing the fourth-generation primary model in the DeepSeek series, positioned as a high-performance open-source reasoning model.
Core Facts
Timeline (last 90 days)
In the LoCoMo evaluation, Hy-Memory 2.0's 72.6% accuracy is higher than GPT-6 Astra's 71.2%, Claude Fable 5.1's 70.8% and DeepSeek 4.1's 68.4%.
In multimodal evaluations, Nemotron 3.5 Super VL goes head-to-head with Claude Fable 5.1, GPT-6 Astra, DeepSeek 4.1, and Gemini 3.8 Flash Cyber, scoring 77.4% on Video-MME and 94.6% on DocVQA.
On Terminal-Bench 4.0, Claude Opus 5.5 leads with 66.4%, followed by GPT-6 Astra at 61.8% and Argon at 57.4%.
On the Intelligence Index, Haiku 5.5 surpassed Gemini 3.8 Flash and DeepSeek 4.1.
All Facts (4)
On Terminal-Bench 4.0, Claude Opus 5.5 leads with 66.4%, followed by GPT-6 Astra at 61.8% and Argon at 57.4%.
75%UnverifiedIn the LoCoMo evaluation, Hy-Memory 2.0's 72.6% accuracy is higher than GPT-6 Astra's 71.2%, Claude Fable 5.1's 70.8% and DeepSeek 4.1's 68.4%.
50%UnverifiedIn multimodal evaluations, Nemotron 3.5 Super VL goes head-to-head with Claude Fable 5.1, GPT-6 Astra, DeepSeek 4.1, and Gemini 3.8 Flash Cyber, scoring 77.4% on Video-MME and 94.6% on DocVQA.
50%UnverifiedOn the Intelligence Index, Haiku 5.5 surpassed Gemini 3.8 Flash and DeepSeek 4.1.
50%