Can CodeGeeX 4 9B run on Intel Arc A380 6GB?

YES — With NVFP4

B59Good

Estimated from fit model

CodeGeeX 4 9B needs ~7.2 GB VRAM. Intel Arc A380 6GB has 6.0 GB. With NVFP4 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.

CodeGeeX 4 9B at Q4_K_M needs 7.6 GB — too much for Intel Arc A380 6GB (6.0 GB). Runs at NVFP4 (7.2 GB) with medium quality. 3 quantization levels fit.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 7.6 GB, exceeds 6.0 GB available

7.6 GB required6.0 GB available

127% VRAM needed

1.6 GB over capacity — needs offload or smaller quantization

Fit status

Too heavy

Decode

8.3 tok/s

TTFT

23385 ms

Safe context

Memory

7.6 GB / 6.0 GB

Offload

20%

Memory breakdown

Weights5.5 GB

KV Cache0.6 GB

Runtime0.9 GB

Headroom0.6 GB

See how fast it feels

With memory offload — actual speed may be lower

See how fast it feelsCodeGeeX 4 9B on Intel Arc A380 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: 8.3 tok/s decode · 23.4s TTFT (warm) · 21 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 20% 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	9.0 tok/s	11701 ms	4K
Coding	F	Too heavy	8.3 tok/s	23385 ms	4K
Agentic Coding	F	Too heavy	7.0 tok/s	40020 ms	4K
Reasoning	F	Too heavy	8.3 tok/s	27637 ms	4K
RAG	F	Too heavy	7.0 tok/s	50025 ms	4K

Inference speed

CodeGeeX 4 9B inference speed — tokens per second by GPU & Mac

Estimated decode speed (tokens/sec) for CodeGeeX 4 9B at Q4_K_M across popular GPUs and Apple Silicon, using the fastest local runtime per device. Fastest is RTX 5090 32GB at ~126 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	126.0	Fits
RTX 4090 24GB	24 GB	Q4_K_M	126.0	Fits
RTX 3090 24GB	24 GB	Q4_K_M	126.0	Fits
RX 7900 XTX 24GB	24 GB	Q4_K_M	126.0	Fits
RTX 4080 Super 16GB	16 GB	Q4_K_M	121.7	Fits
Mac Studio M3 Ultra 256GB	256 GB	Q4_K_M	111.0	Fits
Mac Studio M2 Ultra 128GB	128 GB	Q4_K_M	92.4	Fits
Mac Studio M1 Ultra 128GB	128 GB	Q4_K_M	87.7	Fits
RTX 4070 12GB	12 GB	Q4_K_M	75.3	Fits
MacBook Pro M4 Max 128GB	128 GB	Q4_K_M	74.7	Fits
MacBook Pro M4 Max 64GB	64 GB	Q4_K_M	74.7	Fits
MacBook Pro M3 Max 64GB	64 GB	Q4_K_M	47.8	Fits
RTX 3060 12GB	12 GB	Q4_K_M	47.3	Fits
MacBook Pro M1 Max 64GB	64 GB	Q4_K_M	43.8	Fits
MacBook Pro M4 Pro 48GB	48 GB	Q4_K_M	38.5	Fits
RTX 4060 8GB	8 GB	Q4_K_M	36.4	Offloads

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 CodeGeeX 4 9B (9B params) fits at each quantization level on Intel Arc A380 6GB (6.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_KBest for your GPU	2	3.5 GB	Low	A81
Q3_K_S	3	4.4 GB	Low	F0
NVFP4	4	5.0 GB	Medium	F0
Q4_K_M	4	5.5 GB	Medium	F0
Q5_K_M	5	6.5 GB	High	F0
Q6_K	6	7.4 GB	High	F0
Q8_0	8	9.6 GB	Very High	F0
F16	16	18.5 GB	Maximum	F0

Get started

Copy-paste commands to run CodeGeeX 4 9B on your machine.

Run

docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
  --hf-repo "THUDM/codegeex4-all-9b" \
  --hf-file "codegeex4-all-9b-Q4_K_M.gguf" \
  -c 4096 -ngl 99

Upgrade-Optionen

Hardware, die CodeGeeX 4 9B gut ausführt

Intel Arc A580 8GBGünstige Wahl

8 GB VRAM (+2)512 GB/s (+326)

Makes the model fit on the accelerator instead of staying completely out of reach.50 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 10GBBestes Preis-Leistungs-Verhältnis

10 GB VRAM (+4)380 GB/s (+194)

Makes the model fit on the accelerator instead of staying completely out of reach.40.9 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

Intel Arc B580 12GBIntel-Upgrade

12 GB VRAM (+6)456 GB/s (+270)

Makes the model fit on the accelerator instead of staying completely out of reach.43.6 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. $249 MSRP

Frequently asked questions

Can Intel Arc A380 6GB run CodeGeeX 4 9B?

Yes, Intel Arc A380 6GB can run CodeGeeX 4 9B at NVFP4 quantization (Very compromised (needs ~0.8 GB host RAM)). The recommended Q4_K_M requires 7.6 GB which exceeds available memory, but at NVFP4 it needs only 7.2 GB. Expected decode speed: 10.8 tok/s.

How much VRAM does CodeGeeX 4 9B need?

CodeGeeX 4 9B (9B parameters) requires approximately 7.6 GB at Q4_K_M quantization. On Intel Arc A380 6GB, it fits at NVFP4 using 7.2 GB.

What is the best quantization for CodeGeeX 4 9B?

The recommended quantization is Q4_K_M, but on Intel Arc A380 6GB the best fitting quantization is NVFP4, which uses 7.2 GB.

What speed will CodeGeeX 4 9B run at on Intel Arc A380 6GB?

On Intel Arc A380 6GB, CodeGeeX 4 9B achieves approximately 10.8 tokens per second decode speed with a time-to-first-token of 17981ms using NVFP4 quantization.

Can Intel Arc A380 6GB run CodeGeeX 4 9B for coding?

For coding workloads, CodeGeeX 4 9B on Intel Arc A380 6GB receives a F grade with 8.3 tok/s and 4K context.

What context window can CodeGeeX 4 9B use on Intel Arc A380 6GB?

On Intel Arc A380 6GB, CodeGeeX 4 9B can safely use up to 4K tokens of context at NVFP4 quantization. The model's official context limit is 131K, but available memory constrains the safe maximum.

What should I upgrade first if CodeGeeX 4 9B feels slow on Intel Arc A380 6GB?

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 A380 6GB for CodeGeeX 4 9B?

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 A380 6GB See all hardware for CodeGeeX 4 9B

Embed this result▼

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

<iframe src="https://willitrunai.com/embed/codegeex-4-9b-on-arc-a380-6gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: