Alibaba

Qwen2.5 32B

32,8B

Parameters

18,5 Go

Q4_K_M file

≥ 20,0 Go

Recommended VRAM

32 768

Max context

Not measured yet

No card has run the TokenBench protocol on this model yet. The current protocol covers a selection of models; benchmarking any catalog model from the app is coming soon.

Measure my card →

Which cards should it fit on?

Cards on the site whose video memory covers the recommended 20,0 Go (file + 4,000-token context). An estimate, not a measurement. Click a card for the detailed estimate.

Technical sheet

Published: 2024-09-17
Architecture: 64 layers · 8 KV heads · dimension 128
Attention: Standard: every layer keeps the whole context in memory.
Context cache (f16): ≈ 250 MB per 1,000 tokens · 8,0 Go for 32,768 tokens
GGUF file: bartowski/Qwen2.5-32B-Instruct-GGUF · Qwen2.5-32B-Instruct-Q4_K_M.gguf Download (18,5 Go)
Official repo: Qwen/Qwen2.5-32B-Instruct

Public Hugging Face data (official config.json, GGUF repo). The context cache only counts full-attention layers.