Llama 4 Scout 17B (16E)
109B MoEMeta Llama 4
Meta Llama 4 Scout MoE (17B active / 109B total). Multimodal; needs ~68GB VRAM at Q4_K_M.
10486K
Max Context
3
Quant Variants
GGUF Q4_K_M
Best Quality
97.6%
Accuracy Retained
Quantization Variants
Per-quant VRAM, quality loss, and inference speed on RTX 4090
Measured = site benchmarks · Estimated = formula · Community = public reports
Similar models
Compare with Llama 4Llama 4 Maverick 17B (128E)
Meta Llama 4
Llama 4 Maverick flagship MoE (17B active / 400B total). Multi-GPU or H100 cluster territory.
Mistral Large 3 675B Instruct
Mistral AI
Mistral 3 flagship MoE (41B active / 675B total) with vision encoder. FP8 on 8×H200; GGUF quant for research clusters only.
Llama 3.1 70B Instruct
Meta Llama 3.1
Meta's frontier 70B model. Requires 40GB+ VRAM; dual 3090 or M2 Ultra.
Qwen2.5 72B Instruct
Alibaba Qwen2.5
Flagship Qwen2.5. Requires dual 4090 or A100 80G. Exceptional reasoning at scale.
Running Llama 4 Scout 17B (16E) locally
At Q4_K_M and 4K of context, Llama 4 Scout 17B (16E) needs about 69.9 GB — 62.8 GB of weights, 0.75 GB of KV cache and a 6.3 GB activation buffer. The smallest card in this index that clears that comfortably is the A100 80G at 80 GB, and 10 of the 63 cards here do. These are calculated figures, not measurements: the estimate stops counting a card as comfortable at 88% of its VRAM, which is roughly the room a desktop session needs.
What longer context costs
Going from 4K to 32K adds about 5.25 GB, taking the total to 75.7 GB. Weights do not move with context — only the KV cache does, and it grows linearly, so this is the number to watch when planning for long documents. The model's native window is 10240K; holding all of it at Q4_K_M would need about 2181 GB.
Which build to download
This index tracks 2 formats for it — GGUF, AWQ — across 3 levels. Q4_K_M carries the lowest published perplexity loss at 2.4%. The fastest level measured here is AWQ INT4 at 28 tok/s on an RTX 4090, batch 1. GGUF runs on llama.cpp and Ollama across NVIDIA, AMD and Apple silicon; AWQ and GPTQ target vLLM on CUDA and ROCm; EXL2 is ExLlamaV2 and CUDA only.
Common questions
- How much VRAM does Llama 4 Scout 17B (16E) need?
- About 69.9 GB at Q4_K_M with 4K of context and batch 1, rising to roughly 75.7 GB at 32K. That figure is weights plus KV cache plus a 10% activation buffer, calculated from the model's architecture rather than measured on a card.
- Will Llama 4 Scout 17B (16E) run on a 80GB GPU?
- Yes — at Q4_K_M and 4K context it needs about 69.9 GB, which leaves 10.1 GB spare on a A100 80G. That is the smallest card in this index that clears it comfortably; 10 of 63 do.
- Which quantization of Llama 4 Scout 17B (16E) should I use?
- Q4_K_M has the lowest published quality loss (2.4%), and Q4_K_M is the level most people run. All 3 levels in the index are Q4_K_M, Q3_K_M, AWQ INT4.
Where this model fits
Sized at Q4_K_M with a 4K context window, smallest card first. Comfortable means the estimate uses at most 88% of the memory.
- Runs comfortably on
- A100 80G80GB · +10.1H100 80G80GB · +10.1Mac M5 Max 128G128GB · +26.1Mac M4 Max 128G128GB · +26.1Mac M3 Max 128G128GB · +26.1
- Tight but possible
- Mac M2 Max 96G96GB
- Compared here
- GGUF vs AWQ
- Try it yourself
- Size it in the calculatorGenerate the run command
How to actually run this
Deployment guides for this model and this class of hardware.
This page's figures change when the model or the runtime does.Last updated 2026-09-23 RSS → /feed.xml