Meta
Llama 4 Maverick 400B-A17B
visionMoE
401,6B
Parameters · 17B active
226,1 Go
Q4_K_M file
≥ 227,6 Go
Recommended VRAM
1 048 576
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 227,6 Go (file + 4,000-token context). An estimate, not a measurement. Click a card for the detailed estimate.
Technical sheet
- Published
- 2025-04-01
- Architecture
- 48 layers · 8 KV heads · dimension 128
- Attention
- Standard: every layer keeps the whole context in memory.
- Context cache (f16)
- ≈ 188 MB per 1,000 tokens · 6,0 Go for 32,768 tokens
- GGUF file
- unsloth/Llama-4-Maverick-17B-128E-Instruct-GGUF · 5 files Download (226,1 Go)
- Official repo
- meta-llama/Llama-4-Maverick-17B-128E-Instruct
Public Hugging Face data (official config.json, GGUF repo). The context cache only counts full-attention layers.