Fetching from the wire…
01
02
03
04
05
06
07
08
Models2026-08-28 · source-backed
One file handles tokenizer, transformer, KV cache, sampling and CPU kernels with no external library doing the interesting parts, producing about a 5.0 GB model file (GitHub). Weights are int8 with FP16 scales, linear-layer inputs dynamically quantized to int8, other activations float32, validated against the Hugging Face reference running Google's unquantized QAT checkpoint in BF16. The author confirms the demo is real time, CPU only. Read it if you want to know what a competent inference runtime is actually doing.
Each link below shares sources, entities, or timing with this story.
Qwen competes with Google / Shared entities / Earlier coverage / Tension
Linked by a graph relationship (Qwen competes with Google); both cover E2B, Gemma, Hugging Face, Weights; earlier E2B coverage from 2026-08-15.
DiffusionGemma benchmarked against Gemma / Shared entities / Earlier coverage / Tension
Linked by a graph relationship (DiffusionGemma benchmarked against Gemma); both cover Gemma, Google, Hugging Face; earlier Gemma coverage from 2026-06-14.
Hugging Face released Ettin Reranker / Shared entities / Same source domain / Shared topic / Earlier coverage
Linked by a graph relationship (Hugging Face released Ettin Reranker); both cover GitHub, Hugging Face; reported by the same outlet (github.com).
Hugging Face criticizes OpenAI / Shared entities / Same source domain / Earlier coverage / Tension
Linked by a graph relationship (Hugging Face criticizes OpenAI); both cover GitHub, Google; reported by the same outlet (github.com).
Hugging Face criticizes OpenAI / Shared entity: GitHub / Same source domain / Shared topic / Earlier coverage / Tension
Linked by a graph relationship (Hugging Face criticizes OpenAI); both cover GitHub; reported by the same outlet (github.com).
Claude Code uses Hugging Face / Shared entities / Earlier coverage
Linked by a graph relationship (Claude Code uses Hugging Face); both cover Gemma, Google, Hugging Face; earlier Gemma coverage from 2026-08-16.
Hugging Face partners with NVIDIA / Shared entities / Earlier coverage
Linked by a graph relationship (Hugging Face partners with NVIDIA); both cover GitHub, Google, Hugging Face; earlier GitHub coverage from 2026-07-27.
Linked by a graph relationship (Hugging Face partners with NVIDIA); both cover Gemma, Google, Hugging Face; earlier Gemma coverage from 2026-06-11.