model roundup

DeepSeek 4

4 items · started 2026-06-10 · closed 2026-06-13

  1. I just ran into a bizarre hallucination with Fable 5 Max regarding file analysis. i uploaded several PDF to Fable 5 Max, and out of two of it claude completely refused to process it, claiming the files was password-protected.

  2. Two numbers on this model that don't sit comfortably with each other. The Pro config posts coding scores near the top of every board, 80.6 on SWE-bench Verified and 93.5 on LiveCodeBench.

  3. Conventional LLMs keep the full KV cache loaded during decoding, causing a severe GPU memory bottleneck for ultra-long context serving. In this report, we propose Lookahead Sparse Attention (LSA), a novel inference paradigm powered by a Ne…

  4. Hello, Ive ran out of Pro+, is it possible to use DS4 in cursor ide? thanks

← all threads