Qwen3.8-27B-FP8

Status: placeholder. There are no weights in this repository yet.

Qwen/Qwen3.8-27B has not been released at the time of writing. This repo is a reserved skeleton for a planned quantisation; it exists so the naming and the intended recipe are public up front. Do not attempt to download or serve this repo — there is nothing here but this card. When the base model ships and the quantisation has been built and measured, this card will be replaced with real weights, a recipe, and KLD/perplexity numbers.

Planned build

• Base model: Qwen/Qwen3.8-27B (unreleased)
• Scheme: FP8_DYNAMIC (W8A8), attention kept in bf16
• Tool: llm-compressor, compressed-tensors format
• Recipe: recipes/fp8_dynamic_attnbf16.yaml in the pipeline repo

FP8 dynamic per-token activation quantisation of the LLM trunk's MLPs. The vision tower, the linear_attn SSM block, the entire self_attn block and the MTP head stay in bf16.

What is not yet known

The recipe above is the intended configuration, carried over from our Qwen3.6-27B-family builds. It is contingent on Qwen/Qwen3.8-27B having a comparable architecture. If the release turns out to be mixture-of-experts, to use a different module layout, or to ship without the multi-token-prediction head, the recipe will change and this card will say so.

No quality numbers are claimed here. Every published build in this line ships with per-token KL divergence against the bf16 base and wikitext-2-raw perplexity, measured on our own harness; those will be added when they exist.

Provenance

Built by huginnfork. This is a plain quantisation — no abliteration, no fine-tuning, no merging. Weights derive solely from the upstream base model, whose license and terms apply.