Can Codestral Mamba 7B run on GTX 1660 Ti 6GB?

YES — With Offload

A75Great

Estimated from fit model

Codestral Mamba 7B needs ~6.3 GB VRAM. GTX 1660 Ti 6GB has 6.0 GB. With Q4_K_M quantization, expect ~28 tok/s.

Runtime: llama.cppCapacity: OffloadBandwidth: LowStack: 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) — 6.3 GB, 28.4 tok/s, Runs with offload (needs ~0.2 GB host RAM)

6.3 GB required6.0 GB available

105% VRAM needed

0.3 GB over capacity — needs offload or smaller quantization

Fit status

Runs with offload (needs ~0.2 GB host RAM)

Decode

28.4 tok/s

TTFT

6825 ms

Safe context

Memory

6.3 GB / 6.0 GB

Memory breakdown

Weights4.3 GB

KV Cache0.5 GB

Runtime0.9 GB

Headroom0.6 GB

See how fast it feels

See how fast it feelsCodestral Mamba 7B on GTX 1660 Ti 6GB
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: 28.4 tok/s decode · 6.8s TTFT (warm) · 71 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.

Older PCIe generation

PCIe 3.0 is workable, but it compounds the penalty when you offload heavily or try to scale across multiple cards.

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 (needs ~0 GB host RAM)	31.0 tok/s	3411 ms	8K
Coding	A	Runs with offload (needs ~0.2 GB host RAM)	28.4 tok/s	6825 ms	8K
Agentic Coding	B	Very compromised (needs ~0.5 GB host RAM)	24.1 tok/s	11708 ms	8K
Reasoning	A	Runs with offload (needs ~0.2 GB host RAM)	28.4 tok/s	8066 ms	8K
RAG	B	Very compromised (needs ~0.5 GB host RAM)	24.1 tok/s	14635 ms	8K

Quantization options

How Codestral Mamba 7B (7B params) fits at each quantization level on GTX 1660 Ti 6GB (6.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	2.7 GB	Low	A79
Q3_K_SBest for your GPU	3	3.4 GB	Low	A79
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

Frequently asked questions

Can GTX 1660 Ti 6GB run Codestral Mamba 7B?

Yes, GTX 1660 Ti 6GB can run Codestral Mamba 7B with a A grade (Runs with offload (needs ~0.2 GB host RAM)). Expected decode speed: 28.4 tok/s.

How much VRAM does Codestral Mamba 7B need?

Codestral Mamba 7B (7B parameters) requires approximately 6.3 GB of memory with Q4_K_M quantization.

What is the best quantization for Codestral Mamba 7B?

The recommended quantization for Codestral Mamba 7B is Q4_K_M, which balances quality and memory efficiency.

What speed will Codestral Mamba 7B run at on GTX 1660 Ti 6GB?

On GTX 1660 Ti 6GB, Codestral Mamba 7B achieves approximately 28.4 tokens per second decode speed with a time-to-first-token of 6825ms using Q4_K_M quantization.

Can GTX 1660 Ti 6GB run Codestral Mamba 7B for coding?

For coding workloads, Codestral Mamba 7B on GTX 1660 Ti 6GB receives a A grade with 28.4 tok/s and 8K context.

What context window can Codestral Mamba 7B use on GTX 1660 Ti 6GB?

On GTX 1660 Ti 6GB, Codestral Mamba 7B can safely use up to 8K tokens of context. 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 GTX 1660 Ti 6GB?

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 GTX 1660 Ti 6GB 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-gtx-1660-ti-6gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview:

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.

Older PCIe generation

PCIe 3.0 is workable, but it compounds the penalty when you offload heavily or try to scale across multiple cards.

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	A	Runs with offload (needs ~0 GB host RAM)	31.0 tok/s	3411 ms	8K
Coding	A	Runs with offload (needs ~0.2 GB host RAM)	28.4 tok/s	6825 ms	8K
Agentic Coding	B	Very compromised (needs ~0.5 GB host RAM)	24.1 tok/s	11708 ms	8K
Reasoning	A	Runs with offload (needs ~0.2 GB host RAM)	28.4 tok/s	8066 ms	8K
RAG	B	Very compromised (needs ~0.5 GB host RAM)	24.1 tok/s	14635 ms	8K

Quant

Bits

VRAM

Quality

Fit

Q2_K

2.7 GB

Low

A79

Q3_K_SBest for your GPU

3.4 GB

Low

A79

NVFP4

3.9 GB

Medium

Q4_K_M

4.3 GB