Can Llama 3.1 405B run on AMD Instinct MI350X 288GB?

YES — With Offload

A84Great

Estimated from fit model

Llama 3.1 405B needs ~284.4 GB VRAM. AMD Instinct MI350X 288GB has 288.0 GB. With Q4_K_M quantization, expect ~26 tok/s.

Runtime: llama.cppCapacity: OffloadBandwidth: HighStack: StandardBottleneck: Balanced

Operating mode

Choose the run profile you care about

Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.

Current mode

Balanced

Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 284.4 GB, 25.9 tok/s, Runs with offload

284.4 GB required288.0 GB available

99% VRAM used

Fit status

Runs with offload

Decode

25.9 tok/s

TTFT

7488 ms

Safe context

23K

Memory

284.4 GB / 288.0 GB

Memory breakdown

Weights247.1 GB

KV Cache7.7 GB

Runtime0.9 GB

Headroom28.8 GB

See how fast it feels

See how fast it feelsLlama 3.1 405B on AMD Instinct MI350X 288GB
1st promptCold start — includes initialization
>Write a Python function to sort a list
Here's a Python function that sorts a list using quicksort:

def quicksort(arr):
    if len(arr) <= 1:
        return arr
    pivot = arr[len(arr) // 2]
    left   = [x for x in arr if x < pivot]
    middle = [x for x in arr if x == pivot]
    right  = [x for x in arr if x > pivot]
    return quicksort(left) + middle + quicksort(right)

# Example usage
numbers = [3, 6, 8, 10, 1, 2, 1]
print(quicksort(numbers))  # [1, 1, 2, 3, 6, 8, 10]

The algorithm picks a pivot, partitions the list into smaller, equal, and larger elements, then recursively sorts each partition and concatenates the results.
2nd promptWarm — model ready, no init overhead
>Now add type hints and a docstring
def quicksort(arr: list[int]) -> list[int]:
    """Sort a list using the quicksort algorithm.

    Args:
        arr: List of integers to sort.

    Returns:
        A new sorted list.
    """
    if len(arr) <= 1:
        return arr
    pivot = arr[len(arr) // 2]
    left   = [x for x in arr if x < pivot]
    middle = [x for x in arr if x == pivot]
    right  = [x for x in arr if x > pivot]
    return quicksort(left) + middle + quicksort(right)

Added type hints for the input and return type, plus a Google-style docstring.
Estimated: 25.9 tok/s decode · 7.5s TTFT (warm) · 65 tok/s prefill

What limits this setup

This setup is broadly balanced for this model.

Very little memory headroom

You can run the model, but there is not much room left for longer context, bigger batches, extra apps, or future model updates.

Best improvement path

Buy headroom, not only minimum fit

A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	A	Runs with offload	25.9 tok/s	4084 ms	23K
Coding	A	Runs with offload	25.9 tok/s	7488 ms	23K
Agentic Coding	A	Runs with offload (needs ~3.5 GB host RAM)	18.8 tok/s	14964 ms	23K
Reasoning	A	Runs with offload	25.9 tok/s	8849 ms	23K
RAG	A	Runs with offload (needs ~3.5 GB host RAM)	18.8 tok/s	18705 ms	23K

Inference speed

Llama 3.1 405B inference speed — tokens per second by GPU & Mac

Estimated decode speed (tokens/sec) for Llama 3.1 405B at Q4_K_M across popular GPUs and Apple Silicon, including multi-GPU rigs, using the fastest local runtime per device. Fastest is RTX 5090 32GB at ~2 tok/s. Speed is memory-bandwidth bound, so cards that fit the whole model in VRAM run far faster than ones that offload to system RAM.

GPU / Mac	Memory	Quant	Speed (tok/s)	Fits?
RTX 5090 32GB	32 GB	Q4_K_M	2.0	Too big
RTX 4090 24GB	24 GB	Q4_K_M	2.0	Too big
RTX 4080 Super 16GB	16 GB	Q4_K_M	2.0	Too big
RTX 3090 24GB	24 GB	Q4_K_M	2.0	Too big
RTX 4070 12GB	12 GB	Q4_K_M	2.0	Too big
RTX 3060 12GB	12 GB	Q4_K_M	2.0	Too big
RTX 4060 8GB	8 GB	Q4_K_M	2.0	Too big
RX 7900 XTX 24GB	24 GB	Q4_K_M	2.0	Too big
MacBook Pro M4 Max 128GB	128 GB	Q4_K_M	2.0	Too big
Mac Studio M3 Ultra 256GB	256 GB	Q4_K_M	2.0	Too big
Mac Studio M2 Ultra 128GB	128 GB	Q4_K_M	2.0	Too big
Mac Studio M1 Ultra 128GB	128 GB	Q4_K_M	2.0	Too big
MacBook Pro M4 Max 64GB	64 GB	Q4_K_M	2.0	Too big
MacBook Pro M3 Max 64GB	64 GB	Q4_K_M	2.0	Too big
MacBook Pro M1 Max 64GB	64 GB	Q4_K_M	2.0	Too big
MacBook Pro M4 Pro 48GB	48 GB	Q4_K_M	2.0	Too big
2× RTX 4090 24GB	48 GB	Q4_K_M	2.0	Too big
2× RTX 3090 24GB	48 GB	Q4_K_M	2.0	Too big
2× RX 7900 XTX 24GB	48 GB	Q4_K_M	2.0	Too big
4× RTX 3060 12GB	48 GB	Q4_K_M	2.0	Too big

Estimates for single-stream decoding at Q4_K_M; real tokens/sec varies with prompt length, context, batch size, and runtime build. Prompt processing (prefill) is faster than the decode figures shown here.

Quantization options

How Llama 3.1 405B (405B params) fits at each quantization level on AMD Instinct MI350X 288GB (288.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	158.0 GB	Low	A82
Q3_K_S	3	198.5 GB	Low	A82
NVFP4Best for your GPU	4	226.8 GB	Medium	A82
Q4_K_M	4	247.1 GB	Medium	F0
Q5_K_M	5	291.6 GB	High	F0
Q6_K	6	332.1 GB	High	F0
Q8_0	8	433.4 GB	Very High	F0
F16	16	830.2 GB	Maximum	F0

Get started

Copy-paste commands to run Llama 3.1 405B on your machine.

Run

ollama run llama3.1:405b

Your hardware

More models your AMD Instinct MI350X 288GB can run

Model	Params	Grade	Decode	Capabilities
Qwen3-Coder 480B A35B Instruct	480B	A	35.3 tok/s

Frequently asked questions

Can AMD Instinct MI350X 288GB run Llama 3.1 405B?

Yes, AMD Instinct MI350X 288GB can run Llama 3.1 405B with a A grade (Runs with offload). Expected decode speed: 25.9 tok/s.

How much VRAM does Llama 3.1 405B need?

Llama 3.1 405B (405B parameters) requires approximately 284.4 GB of memory with Q4_K_M quantization.

What is the best quantization for Llama 3.1 405B?

The recommended quantization for Llama 3.1 405B is Q4_K_M, which balances quality and memory efficiency.

What speed will Llama 3.1 405B run at on AMD Instinct MI350X 288GB?

On AMD Instinct MI350X 288GB, Llama 3.1 405B achieves approximately 25.9 tokens per second decode speed with a time-to-first-token of 7488ms using Q4_K_M quantization.

Can AMD Instinct MI350X 288GB run Llama 3.1 405B for coding?

For coding workloads, Llama 3.1 405B on AMD Instinct MI350X 288GB receives a A grade with 25.9 tok/s and 23K context.

What context window can Llama 3.1 405B use on AMD Instinct MI350X 288GB?

On AMD Instinct MI350X 288GB, Llama 3.1 405B can safely use up to 23K tokens of context. The model's official context limit is 131K, but available memory constrains the safe maximum.

What should I upgrade first if Llama 3.1 405B feels slow on AMD Instinct MI350X 288GB?

Buy headroom, not only minimum fit. A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.

See all results for AMD Instinct MI350X 288GB See all hardware for Llama 3.1 405B

Embed this result▼

Paste this snippet into any page to show a live fit card.

<iframe src="https://willitrunai.com/embed/llama-3.1-405b-on-instinct-mi350x-288gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: