Meta
ChatReasoningMulti-purpose109B
Chat

Llama 4 Scout

Llama · Llama 4 Community License

Mixture of Experts: 109B total parameters, 17B active per token.

Large MoE model (~109B total, ~17B active per token; 16 experts, 2 active). Multimodal (text and image) with an enormous advertised context window (10M tokens); practical local runs are usually capped by VRAM and tooling far below that. MoE loads the full weight set for common local inference paths. Not a consumer GPU model: even Q4_K_M needs ~60GB VRAM. Expect H100-class or dual A100 hardware for practical deployment. Do not recommend to RTX 4090 or Mac users.

Parameters
109B
Architecture
MoE (17B active)
Context
10,000,000 tokens
Released
2025-04-05
Engines
llama.cpp, ollama, vLLM
Builder Tools
Continue, LM Studio, Open WebUI

Parameters

109B

VRAM

75 GB

Context

9766K

Formats

4

GPUs

13

Llama 4 Scout (109B) requires 75 GB VRAM at recommended quality (Q5_K_M). On NVIDIA RTX PRO 6000 Blackwell, expect approximately 95 tok/s at Q5_K_M. For the best experience, Mac Studio AI Builder ($3,999) is recommended.

Source: OwnRig methodology

VRAM (Recommended)

75 GB

Quantization

Q5_K_M

File Size

72 GB

Max Context

9766K tokens

Primary Use

Chat

Memory

VRAM Requirements

QualityQuantizationVRAMFile Size
fullQ8_0110 GB109 GB
recommendedQ5_K_M75 GB72 GB
efficientQ4_K_M60 GB58 GB
compressedQ3_K_M50 GB48 GB
Scaling

Context Length Impact

KV cache VRAM at Q5_K_M quality. Longer context = more memory.

ContextKV CacheTotal VRAM
2K307 MB75.3 GBexceeds 24 GB
4K614 MB75.6 GBexceeds 24 GB
8K1.2 GB76.2 GBexceeds 24 GB
16K2.4 GB77.4 GBexceeds 24 GB
32K4.8 GB79.8 GBexceeds 24 GB
64K9.6 GB84.6 GBexceeds 24 GB
128K19.2 GB94.2 GBexceeds 24 GB

Compatible GPUs

13 devices
NVIDIA Grace Blackwell Ultra GB300Q8_040 tok/sExcellent
NVIDIA RTX PRO 6000 BlackwellQ5_K_M95 tok/sExcellent
NVIDIA RTX PRO 6000 Blackwell Max-QQ5_K_M87 tok/sExcellent
Apple M4 Max (128GB Unified)Q8_04 tok/sMarginal
Apple M4 Max (64GB Unified)Q4_K_M5 tok/sMarginal
Apple M4 Pro (48GB)Q3_K_M6 tok/sMarginal
Apple M4 Ultra (192GB)Q8_05 tok/sMarginal
AMD Radeon Pro W7900Q3_K_M2 tok/sMarginal
AMD Radeon RX 7600Q3_K_MNot viable
AMD Radeon RX 7900 XTXQ3_K_MNot viable
AMD Radeon RX 9070Q3_K_MNot viable
AMD Radeon RX 9060 XT 16GBQ3_K_MNot viable
AMD Radeon RX 9060 XT 8GBQ3_K_MNot viable

Showing 13 of 13 entries

Builder Context

Llama 4 Scout is commonly used with Continue, LM Studio, Open WebUI.

FAQ

Frequently Asked Questions

How much VRAM does Llama 4 Scout need?
Llama 4 Scout requires 75 GB VRAM at recommended quality (Q5_K_M). At lower quality settings, it can fit in as little as 50 GB.
What is the best GPU for Llama 4 Scout?
The NVIDIA RTX PRO 6000 Blackwell delivers the best performance for Llama 4 Scout, achieving 95 tok/s at Q5_K_M with an excellent rating.
What quantization should I use for Llama 4 Scout?
For the best quality, use Q5_K_M (75 GB VRAM). If your GPU has limited VRAM, Q3_K_M (50 GB) is the most efficient option with acceptable quality.
All models

Data confidence: estimated. Source

VRAM requirements are calculated from model parameters and may vary by inference engine, context length, and batch size. Performance estimates are based on community benchmarks and should be verified for your specific configuration.Llama is a trademark of its respective owner. OwnRig is not affiliated with or endorsed by the model creator.