BestLLMfor EN Your hardware. Your LLM. Your call.
APIOpen data Find my LLM
Model fiche

Llama 4 Maverick 400B

By Meta · United States

Updated 2026-07-13

chat general vision moe multilingual
Parameters
400B
License
Llama 4 Community
Context
976k
VRAM (Q4)
240 GB
Released
April 2025

Overview

Meta's larger Llama 4 MoE at 400B total with 17B active across 128 experts, natively multimodal. LMArena 1417 and 1M token context, but 245GB to download.

When to pick this model

  • Frontier-quality open chat in multi-GPU production
  • Multimodal agents needing 1M context
  • Drop-in for teams ready to commit to the Llama 4 ecosystem
  • Workloads where MMLU-Pro 80 quality justifies the storage cost

VRAM requirements by quantization

VRAM REQUIRED (GB)80128256512Q4_K_M240 GBQ5_K_M285 GBQ8_0425 GBFP16800 GB
QuantizationVRAM required
Q4_K_M (recommended)240 GB
Q5_K_M285 GB
Q8_0425 GB
FP16 (no quantization)800 GB

VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.

In practice, Llama 4 Maverick 400B is server-class even at Q4_K_M (240 GB). Stepping up to Q8_0 nearly doubles the footprint to 425 GB, and unquantized FP16 weights take 800 GB — plan your GPU around the Q4 or Q5 figure unless you specifically need the higher fidelity.

Without a GPU, Llama 4 Maverick 400B needs roughly 280 GB of system RAM to run on CPU via llama.cpp or Ollama — workable for background jobs, but far slower than GPU inference. Throughput estimates from our compatibility engine: around 2 tokens/sec on entry-level GPUs, on the order of 8 tokens/sec on a mid-range card, and up to 22 tokens/sec on high-end hardware — assuming the chosen quantization fully fits in VRAM.

What hardware do you need

The table below matches Llama 4 Maverick 400B to common GPU memory tiers, using the highest-fidelity quantization that fully fits each card class. Spilling layers to system RAM works but costs most of the speed, so size your card to the quantization you actually want to run.

GPU memoryExample cardsBest fit for Llama 4 Maverick 400B
8 GBRTX 5070 Laptop, RTX 5060, RTX 5060 Ti 8GBDoes not fit — needs 240 GB at Q4_K_M
12 GBRTX 5070, RTX 5070 Ti Laptop, RTX 4080 LaptopDoes not fit — needs 240 GB at Q4_K_M
16 GBRTX 5080, RTX 4080 Super, Radeon RX 9070 XTDoes not fit — needs 240 GB at Q4_K_M
24 GBRTX 4090, Radeon RX 7900 XTX, RTX 5090 LaptopDoes not fit — needs 240 GB at Q4_K_M
32 GBRTX 5090Does not fit — needs 240 GB at Q4_K_M

Which GPU should you buy to run Llama 4 Maverick 400B?

To run Llama 4 Maverick 400B locally at Q4, you need ~240 GB of VRAM. The best value for this is a Apple Mac Studio (64+ GB unified memory).

Check Apple Mac Studio price on Amazon →

As an Amazon Associate, BestLLMfor earns from qualifying purchases, at no extra cost to you. It does not influence our independent rankings.

Published benchmark scores

BenchmarkScore
LMArena70.85
MMLU-Pro80

Scores published by the model author or aggregated from public leaderboards. Re-measured monthly by our editorial team.

Strengths

  • LMArena 1417 — top-tier open chat quality
  • MMLU-Pro 80
  • 1M token context
  • Native multimodal with strong vision performance
  • 17B active keeps inference cost manageable

Limitations

  • 245GB download — non-trivial storage and bandwidth
  • Hugging Face gated access
  • Llama 4 Community License with >700M MAU clause
  • Outclassed on reasoning by R1-class models

Typical workloads

In our catalog grid, Llama 4 Maverick 400B is filed under Multimodal Frontier, Agents, Vision — the use cases where its size/quality trade-off makes the most sense. Its tags translate to concrete workloads: vision-language work — screenshots, charts, scanned documents; multilingual workloads.

The 976k-token context window is large enough to hold entire codebases' worth of files or long reports in a single prompt, which is what makes local RAG and document analysis practical. It ships under the Llama 4 Community license — commercial use is generally possible but read the specific terms before embedding it in a product.

Architecture & training

Architecture: MoE 128 experts · 400B/17B active · natively multimodal · 1M ctx

Training: Scout's big brother.

Verdict

Meta's biggest open chat model and a credible GPT-4-class alternative — if you can host 245GB and accept the MAU clause.

Quick start

ollama run llama4:maverick

Or use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.

Similar models worth comparing

Frequently asked questions

How much VRAM does Llama 4 Maverick 400B need?

At the recommended Q4_K_M quantization, Llama 4 Maverick 400B needs about 240 GB of VRAM. Q8_0 takes 425 GB, and unquantized FP16 weights take 800 GB.

Can Llama 4 Maverick 400B run without a GPU?

Yes — with roughly 280 GB of system RAM it runs CPU-only through llama.cpp or Ollama. Expect a fraction of GPU speed, which is fine for background or batch jobs but slow for interactive chat.

What context window does Llama 4 Maverick 400B support?

Llama 4 Maverick 400B supports a 976k-token context window (1,000,000 tokens).

Can I use Llama 4 Maverick 400B commercially?

Llama 4 Maverick 400B ships under the Llama 4 Community license. Commercial use is generally permitted subject to its terms — review the license text before shipping a product.

How fast is Llama 4 Maverick 400B on consumer hardware?

Our compatibility engine estimates on the order of 8 tokens/sec on a mid-range GPU and up to 22 tokens/sec on high-end cards, assuming the quantization fully fits in VRAM.

Which quantization of Llama 4 Maverick 400B should I download first?

Start with Q4_K_M (240 GB) — the standard size/quality sweet spot. Step up to Q5_K_M or Q8_0 only if you have VRAM headroom. It does not fit a single 24 GB consumer card — plan for multi-GPU or server hardware.

Tools

Is Llama 4 Maverick 400B the right pick for you?

Compute self-hosted ROI → Back to catalog