Llama 4 Maverick 17B — VRAM, speed & local setup
Written by Jakub Rusinowski · Last updated
The premium Llama 4 model. 128 experts with 17B active parameters. Matches GPT-4o in benchmarks. Its 400B total parameters must all be resident, which is roughly 242 GB at Q4_K_M — a multi-GPU server or a 256 GB+ unified-memory system. The 17B active count sets its speed, not its memory footprint.
Llama 4 Maverick 17B needs about 242 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache. On Apple Silicon that figure comes out of unified memory.
VRAM and speed by quantization
Quoted against NVIDIA RTX 4090 (24 GB). Includes the KV cache at 8K context, so it reads higher than the headline figure.
| Quant | Memory | VRAM | Speed (est.) | Fit |
|---|---|---|---|---|
| Q2_K 2.63 bpw | 133.9 GB | — | Too big | |
| Q3_K_M 3.41 bpw | 172.9 GB | — | Too big | |
| Q4_K_M 4.83 bpw | 243.9 GB | — | Too big | |
| Q5_K_M 5.67 bpw | 285.9 GB | — | Too big | |
| Q6_K 6.56 bpw | 330.4 GB | — | Too big | |
| Q8_0 8.50 bpw | 427.4 GB | — | Too big | |
| F16 16.00 bpw | 802.4 GB | — | Too big |
Black marker = usable memory on the NVIDIA RTX 4090 (24 GB). Estimates from the memory-bandwidth roofline on the methodology page. Llama 4 Maverick 17B VRAM calculator →
Get Llama 4 Maverick 17B running
The cheapest catalogued GPU that runs Llama 4 Maverick 17B is the Apple M3 Ultra (512 GB).
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
How to run Llama 4 Maverick 17B
Install Ollama, then run:
ollama run llama4:maverickSpecifications
Curated — A hand-written entry from before this catalogue recorded its sources. The figures are long-standing but their provenance is not on file.
- Parameters
- 400 Billion (17B active)
- Context window
- 1,000,000
- Architecture
- MoE (128 experts)
- Provider
- Meta
- Licence
- Llama 4 Community
- Practical deployment
- Workstation (large unified memory or pro GPU)
- Specified at
- Q4_K_M
- System RAM
- 32 GB
- Record updated
- 2025-04-05
Weights are downloadable and commercial use is permitted, subject to the licence’s acceptable-use terms.
Quality and use cases
Scores as published by the model’s authors or an independent evaluator — quality, not throughput, and not measured by us.
Can I run Llama 4 Maverick 17B on my GPU?
- Llama 4 Maverick 17B on AMD Ryzen AI Max+ 395
- Llama 4 Maverick 17B on Apple M1 Max
- Llama 4 Maverick 17B on Apple M1 Pro
- Llama 4 Maverick 17B on Apple M1 Ultra
- Llama 4 Maverick 17B on Apple M2 Max
- Llama 4 Maverick 17B on Apple M2 Pro
- Llama 4 Maverick 17B on Apple M3 Max
- Llama 4 Maverick 17B on Apple M3 Pro
- Llama 4 Maverick 17B on Apple M3 Ultra
- Llama 4 Maverick 17B on Apple M4
- Llama 4 Maverick 17B on Apple M4 Max
- Llama 4 Maverick 17B on Apple M5
- Llama 4 Maverick 17B on Apple M5 Max
- Llama 4 Maverick 17B on Apple M5 Pro
- Llama 4 Maverick 17B on NVIDIA A100 80GB (PCIe)
- Llama 4 Maverick 17B on NVIDIA DGX Spark
- Llama 4 Maverick 17B on NVIDIA GeForce RTX 4090
- Llama 4 Maverick 17B on NVIDIA GeForce RTX 5090
- Llama 4 Maverick 17B on NVIDIA H100 80GB (PCIe)
- Llama 4 Maverick 17B on NVIDIA L40S
- Llama 4 Maverick 17B on NVIDIA RTX 6000 Ada Generation
- Llama 4 Maverick 17B on NVIDIA RTX PRO 6000 Blackwell
Other Llama 4 sizes
Llama 4 Maverick 17B — frequently asked questions
How much VRAM does Llama 4 Maverick 17B need?
About 242 GB at Q4_K_M — quantized weights plus framework overhead, before any KV cache. The cache grows with context length and is added on top; the table above folds it in. Apple Silicon counts unified memory toward the same figure.
Does Llama 4 Maverick 17B run on an RTX 4090 (24 GB)?
No. Llama 4 Maverick 17B needs about 242 GB at Q4_K_M, more than a single RTX 4090's 24 GB. It needs a larger card, several GPUs, or Apple Silicon with enough unified memory — or it runs with part of the weights offloaded to system RAM, which is much slower.
How do I run Llama 4 Maverick 17B locally?
Install Ollama and run `ollama run llama4:maverick`. That pulls the weights and starts a local OpenAI-compatible endpoint; after the download nothing leaves the machine.
What other sizes does Llama 4 come in?
Llama 4 Scout 17B (67 GB), Llama 4 Maverick 17B (242 GB). Every size shares the family's training and licence; the larger ones score higher and need proportionally more memory.