Can Codestral Mamba 7B run on Intel Arc A370M 4GB?

YES — With Q2_K

B62Good

Estimated from fit model

Codestral Mamba 7B needs ~4.5 GB VRAM. Intel Arc A370M 4GB has 4.0 GB. With Q2_K quantization, expect ~11 tok/s.

Runtime: llama.cppCapacity: OffloadBandwidth: Very lowStack: StandardBottleneck: Host offload

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.

Codestral Mamba 7B at Q4_K_M needs 6.1 GB — too much for Intel Arc A370M 4GB (4.0 GB). Runs at Q2_K (4.5 GB) with low quality.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 6.1 GB, exceeds 4.0 GB available

6.1 GB required4.0 GB available

153% VRAM needed

2.1 GB over capacity — needs offload or smaller quantization

Fit status

Too heavy

Decode

4.6 tok/s

TTFT

41847 ms

Safe context

Memory

6.1 GB / 4.0 GB

Offload

30%

Memory breakdown

Weights4.3 GB

KV Cache0.5 GB

Runtime0.9 GB

Headroom0.4 GB

See how fast it feels

With memory offload — actual speed may be lower

See how fast it feelsCodestral Mamba 7B on Intel Arc A370M 4GB
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: 4.6 tok/s decode · 41.8s TTFT (warm) · 12 tok/s prefill

What limits this setup

It fits through host-memory offload, and offload is the main reason performance drops.

CPU or host-memory offload is active

About 10% of the working set spills out of accelerator memory, which usually hurts latency and sustained decode throughput.

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.

Runtime ecosystem is narrower than CUDA

Intel GPUs can look attractive on memory per dollar, but local AI tooling, kernels, and model coverage are still broader and easier on CUDA today.

Best improvement path

Remove offload with more accelerator memory

Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.

Prefer CUDA if you want the path of least resistance

If your goal is maximum runtime coverage, easier troubleshooting, and better support for new local AI releases, CUDA is usually still the safer upgrade 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	F	Too heavy	5.0 tok/s	20933 ms	4K
Coding	F	Too heavy	4.6 tok/s	41847 ms	4K
Agentic Coding	F	Too heavy	3.9 tok/s	71657 ms	4K
Reasoning	F	Too heavy	4.6 tok/s	49456 ms	4K
RAG	F	Too heavy	3.9 tok/s	89571 ms	4K

Quantization options

How Codestral Mamba 7B (7B params) fits at each quantization level on Intel Arc A370M 4GB (4.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	2.7 GB	Low	F0
Q3_K_S	3	3.4 GB	Low	F0
NVFP4	4	3.9 GB	Medium	F0
Q4_K_M	4	4.3 GB	Medium	F0
Q5_K_M	5	5.0 GB	High	F0
Q6_K	6	5.7 GB	High	F0
Q8_0	8	7.5 GB	Very High	F0
F16	16	14.3 GB	Maximum	F0

Get started

Copy-paste commands to run Codestral Mamba 7B on your machine.

Run

docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
  --hf-repo "mistralai/Mamba-Codestral-7B-v0.1" \
  --hf-file "Mamba-Codestral-7B-v0.1-Q4_K_M.gguf" \
  -c 4096 -ngl 99

Upgrade-Optionen

Hardware, die Codestral Mamba 7B gut ausführt

Intel Arc A380 6GBGünstige Wahl

6 GB VRAM (+2)186 GB/s (+74)

Makes the model fit on the accelerator instead of staying completely out of reach.16.8 tok/s Dekodierung

Makes the model fit on the accelerator instead of staying completely out of reach.

Removes host-memory offload, which is usually the single biggest latency and throughput win.

ca. $139 MSRP

Intel Arc A580 8GBBestes Preis-Leistungs-Verhältnis

8 GB VRAM (+4)512 GB/s (+400)

Makes the model fit on the accelerator instead of staying completely out of reach.67.6 tok/s Dekodierung

Makes the model fit on the accelerator instead of staying completely out of reach.

Removes host-memory offload, which is usually the single biggest latency and throughput win.

ca. $179 MSRP

Intel Arc B570 10GBIntel-Upgrade

10 GB VRAM (+6)380 GB/s (+268)

Makes the model fit on the accelerator instead of staying completely out of reach.55.3 tok/s Dekodierung

Makes the model fit on the accelerator instead of staying completely out of reach.

Removes host-memory offload, which is usually the single biggest latency and throughput win.

ca. $219 MSRP

Frequently asked questions

Can Intel Arc A370M 4GB run Codestral Mamba 7B?

Yes, Intel Arc A370M 4GB can run Codestral Mamba 7B at Q2_K quantization (Very compromised (needs ~0.3 GB host RAM)). The recommended Q4_K_M requires 6.1 GB which exceeds available memory, but at Q2_K it needs only 4.5 GB. Expected decode speed: 11.4 tok/s.

How much VRAM does Codestral Mamba 7B need?

Codestral Mamba 7B (7B parameters) requires approximately 6.1 GB at Q4_K_M quantization. On Intel Arc A370M 4GB, it fits at Q2_K using 4.5 GB.

What is the best quantization for Codestral Mamba 7B?

The recommended quantization is Q4_K_M, but on Intel Arc A370M 4GB the best fitting quantization is Q2_K, which uses 4.5 GB.

What speed will Codestral Mamba 7B run at on Intel Arc A370M 4GB?

On Intel Arc A370M 4GB, Codestral Mamba 7B achieves approximately 11.4 tokens per second decode speed with a time-to-first-token of 16977ms using Q2_K quantization.

Can Intel Arc A370M 4GB run Codestral Mamba 7B for coding?

For coding workloads, Codestral Mamba 7B on Intel Arc A370M 4GB receives a F grade with 4.6 tok/s and 4K context.

What context window can Codestral Mamba 7B use on Intel Arc A370M 4GB?

On Intel Arc A370M 4GB, Codestral Mamba 7B can safely use up to 4K tokens of context at Q2_K quantization. The model's official context limit is 262K, but available memory constrains the safe maximum.

What should I upgrade first if Codestral Mamba 7B feels slow on Intel Arc A370M 4GB?

Remove offload with more accelerator memory. Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.

Would CUDA be a better path than Intel Arc A370M 4GB for Codestral Mamba 7B?

Often yes, if your goal is the easiest setup and the widest runtime support. Intel can offer attractive memory capacity, but CUDA still tends to win on tooling maturity, guides, kernels, and model coverage for local AI.

See all results for Intel Arc A370M 4GB See all hardware for Codestral Mamba 7B

Embed this result▼

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

<iframe src="https://willitrunai.com/embed/codestral-mamba-7b-on-arc-a370m-4gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: