Can Qwen 3.5 35B A3B run on AMD Instinct MI250X 128GB?

YES — Runs Great

S90Excellent

Estimated from fit model

Qwen 3.5 35B A3B needs ~36.5 GB VRAM. AMD Instinct MI250X 128GB has 128.0 GB. With Q4_K_M quantization, expect ~345 tok/s.

Runtime: llama.cppCapacity: RoomyBandwidth: 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) — 36.5 GB, 344.9 tok/s, Runs well

36.5 GB required128.0 GB available

29% VRAM used

Fit status

Runs well

Decode

344.9 tok/s

TTFT

561 ms

Safe context

131K

Memory

36.5 GB / 128.0 GB

Memory breakdown

Weights21.3 GB

KV Cache1.5 GB

Runtime0.9 GB

Headroom12.8 GB

See how fast it feels

See how fast it feelsQwen 3.5 35B A3B on AMD Instinct MI250X 128GB
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: 344.9 tok/s decode · 561ms TTFT (warm) · 862 tok/s prefill

What limits this setup

This setup is broadly balanced for this model.

No major red flags

This recommendation has enough memory headroom and acceptable estimated speed for the selected workload.

Best improvement path

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	S	Runs well	344.9 tok/s	350 ms	131K
Coding	S	Runs well	344.9 tok/s	561 ms	131K
Agentic Coding	S	Runs well	344.9 tok/s	816 ms	131K
Reasoning	S	Runs well	344.9 tok/s	663 ms	131K
RAG	S	Runs well	344.9 tok/s	1020 ms	131K

Inference speed

Qwen 3.5 35B A3B inference speed — tokens per second by GPU & Mac

Estimated decode speed (tokens/sec) for Qwen 3.5 35B A3B at Q4_K_M across popular GPUs and Apple Silicon, using the fastest local runtime per device. Fastest is RTX 5090 32GB at ~139 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	139.4	Tight
Mac Studio M3 Ultra 256GB	256 GB	Q4_K_M	100.1	Fits
Mac Studio M2 Ultra 128GB	128 GB	Q4_K_M	83.4	Fits
Mac Studio M1 Ultra 128GB	128 GB	Q4_K_M	79.1	Fits
RTX 4090 24GB	24 GB	Q4_K_M	71.1	Heavy offload
MacBook Pro M4 Max 128GB	128 GB	Q4_K_M	61.8	Fits
MacBook Pro M4 Max 64GB	64 GB	Q4_K_M	61.8	Fits
RX 7900 XTX 24GB	24 GB	Q4_K_M	60.0	Heavy offload
RTX 3090 24GB	24 GB	Q4_K_M	56.9	Heavy offload
MacBook Pro M3 Max 64GB	64 GB	Q4_K_M	47.4	Fits
MacBook Pro M1 Max 64GB	64 GB	Q4_K_M	43.5	Fits
MacBook Pro M4 Pro 48GB	48 GB	Q4_K_M	41.5	Tight
RTX 4080 Super 16GB	16 GB	Q4_K_M	25.2	Too big
RTX 4070 12GB	12 GB	Q4_K_M	8.8	Too big
RTX 3060 12GB	12 GB	Q4_K_M	5.2	Too big
RTX 4060 8GB	8 GB	Q4_K_M	4.4	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 Qwen 3.5 35B A3B (35B params) fits at each quantization level on AMD Instinct MI250X 128GB (128.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	13.7 GB	Low	A80
Q3_K_S	3	17.2 GB	Low	A80
NVFP4	4	19.6 GB	Medium	A81
Q4_K_M	4	21.3 GB	Medium	A81
Q5_K_M	5	25.2 GB	High	A81
Q6_K	6	28.7 GB	High	A82
Q8_0	8	37.5 GB	Very High	A83
F16Best for your GPU	16	71.8 GB	Maximum	S89

Get started

Copy-paste commands to run Qwen 3.5 35B A3B on your machine.

Run

ollama run qwen3.5:35b-a3b

Your hardware

More models your AMD Instinct MI250X 128GB can run

Model	Params	Grade	Decode	Capabilities
Devstral 2 123B Instruct	123B	S	36.2 tok/s
Qwen 3.5 122B A10B	122B	S	100.3 tok/s

Frequently asked questions

Can AMD Instinct MI250X 128GB run Qwen 3.5 35B A3B?

Yes, AMD Instinct MI250X 128GB can run Qwen 3.5 35B A3B with a S grade (Runs well). Expected decode speed: 344.9 tok/s.

How much VRAM does Qwen 3.5 35B A3B need?

Qwen 3.5 35B A3B (35B parameters) requires approximately 36.5 GB of memory with Q4_K_M quantization.

What is the best quantization for Qwen 3.5 35B A3B?

The recommended quantization for Qwen 3.5 35B A3B is Q4_K_M, which balances quality and memory efficiency.

What speed will Qwen 3.5 35B A3B run at on AMD Instinct MI250X 128GB?

On AMD Instinct MI250X 128GB, Qwen 3.5 35B A3B achieves approximately 344.9 tokens per second decode speed with a time-to-first-token of 561ms using Q4_K_M quantization.

Can AMD Instinct MI250X 128GB run Qwen 3.5 35B A3B for coding?

For coding workloads, Qwen 3.5 35B A3B on AMD Instinct MI250X 128GB receives a S grade with 344.9 tok/s and 131K context.

What context window can Qwen 3.5 35B A3B use on AMD Instinct MI250X 128GB?

On AMD Instinct MI250X 128GB, Qwen 3.5 35B A3B can safely use up to 131K tokens of context. The model's official context limit is 131K, but available memory constrains the safe maximum.

See all results for AMD Instinct MI250X 128GB See all hardware for Qwen 3.5 35B A3B

Embed this result▼

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

<iframe src="https://willitrunai.com/embed/qwen-3.5-35b-a3b-on-instinct-mi250x-128gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: