Qwen3-Embedding-4B-FP8-Dynamic

Qwen3-Embedding-4B-FP8-Dynamic

Qwen/Qwen3-Embedding-4B quantised to FP8 with llm-compressor, for serving with vLLM.

What was done

Embeddings are the pooled last hidden state; the head is unused and left unquantised. Cosine similarity is scale-sensitive, so compare rankings rather than raw distances when validating this checkpoint.

Quality gate

WikipediaRetrievalMultilingual from MTEB — retrieval over Wikipedia in 16 languages, scored by ndcg_at_10.

Why

Serving Qwen/Qwen3-Embedding-4B in bf16 leaves little room for KV cache on a small GPU: the weights take what the cache needs, and the context length has to be cut until it fits. Halving the weights gives that memory back — the same card serves a longer context without any other change.

Serving

vllm serve DCC-BS/Qwen3-Embedding-4B-FP8-Dynamic \
  --task embed \
  --max-model-len 32768 \
  --gpu-memory-utilization 0.85

The checkpoint is in compressed-tensors format, so vLLM detects the quantisation from config.json; no extra flag is required.