AI News Update

Daily digest for 2026-11-08 · 5 new stories (previously covered stories are never reposted)

Hugging Face Trending1

xtools-at/DeepSeek-V4-Pro-Qwen3.5-9B-UNCENSORED-heretic
Hugging Face Trending

Simon Willison1

datasette #3046: Don't serve link: rel=next HTTP headers to Firefox for HTML pages
Simon Willison · 2026-11-07
datasette #3046: Don't serve link: rel=next HTTP headers to Firefox for HTML pages I was looking at the Firefox network panel today when I noticed that requests to a Datasette table page were triggering an HTTP prefetch to the next page of rows in that table. This was happening b

llama.cpp Releases3

b12205: webgpu : add COUNT_EQUAL support (#31578)
llama.cpp Releases · 2026-11-08
webgpu : add COUNT_EQUAL support Assisted-by: Codex webgpu : reduce small COUNT_EQUAL inputs in one dispatch Assisted-by: Codex webgpu : simplify COUNT_EQUAL and reuse existing tests webgpu : move COUNT_EQUAL pipeline cache declaration Update ggml/src/ggml-webgpu/ggml-webgpu.cpp
b12204: ci: move runners to Ubuntu 26.04 (#31529)
llama.cpp Releases · 2026-11-08
ci: update to ubuntu26.04 Signed-off-by: Aaron Teo [email protected] ci: bring back toolchain workaround Signed-off-by: Aaron Teo [email protected] ci: remove workaround completely Signed-off-by: Aaron Teo [email protected] ci: more switching to ubuntu 26.04 Signed-off-by: Aar
b12203: ggml : switch to F32 src1 in the CPU backend (#31854)
llama.cpp Releases · 2026-11-08
ggml : switch to F32 src1 in the CPU backend Change the default computation for GGML_TYPE_F16 to use F32 for src1 data in matrix multiplications: vec_dot_type(F16) is now F32 and the dot product goes through the new ggml_vec_dot_f16_f32 (x = F16 src0, y = F32 src1), so src1 is no