Can InternLM 20B run on Mac Studio M2 Ultra 128GB?

YES — Runs Great

B57Good

Estimated from fit model

InternLM 20B needs ~47.4 GB VRAM. Mac Studio M2 Ultra 128GB has 92.2 GB. With Q5_K_M quantization, expect ~33 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

Q5_K_M (High quality) — 47.4 GB, 32.9 tok/s, Runs well

47.4 GB required92.2 GB available

51% VRAM used

Fit status

Runs well

Decode

32.9 tok/s

TTFT

5890 ms

Safe context

Memory

47.4 GB / 92.2 GB

Memory breakdown

Weights14.4 GB

KV Cache18.3 GB

Runtime0.9 GB

Headroom13.8 GB

See how fast it feels

See how fast it feelsInternLM 20B on Mac Studio M2 Ultra 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: 32.9 tok/s decode · 5.9s TTFT (warm) · 82 tok/s prefill

What limits this setup

This setup is broadly balanced for this model.

Shared-memory contention still exists

The OS, browser, and inference runtime all compete for the same physical memory pool, so real-world headroom is less forgiving than raw capacity suggests.

Best improvement path

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	B	Runs well	32.9 tok/s	3213 ms	8K
Coding	B	Runs well	32.9 tok/s	5890 ms	8K
Agentic Coding	B	Runs well	32.9 tok/s	8568 ms	8K
Reasoning	B	Runs well	32.9 tok/s	6961 ms	8K
RAG	B	Runs well	32.9 tok/s	10710 ms	8K

Inference speed

InternLM 20B inference speed — tokens per second by GPU & Mac

Estimated decode speed (tokens/sec) for InternLM 20B at Q5_K_M across popular GPUs and Apple Silicon, using the fastest local runtime per device. Fastest is RTX 5090 32GB at ~45 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	Q5_K_M	44.5	Heavy offload
Mac Studio M3 Ultra 256GB	256 GB	Q5_K_M	39.4	Fits
Mac Studio M2 Ultra 128GB	128 GB	Q5_K_M	32.9	Fits
Mac Studio M1 Ultra 128GB	128 GB	Q5_K_M	31.2	Fits
MacBook Pro M4 Max 128GB	128 GB	Q5_K_M	30.7	Fits
MacBook Pro M4 Max 64GB	64 GB	Q5_K_M	30.7	Tight
MacBook Pro M3 Max 64GB	64 GB	Q5_K_M	17.0	Tight
MacBook Pro M4 Pro 48GB	48 GB	Q5_K_M	16.1	Heavy offload
RX 7900 XTX 24GB	24 GB	Q5_K_M	15.6	Too big
MacBook Pro M1 Max 64GB	64 GB	Q5_K_M	15.6	Tight
RTX 4090 24GB	24 GB	Q5_K_M	14.0	Too big
RTX 3090 24GB	24 GB	Q5_K_M	12.9	Too big
RTX 4080 Super 16GB	16 GB	Q5_K_M	5.3	Too big
RTX 4070 12GB	12 GB	Q5_K_M	3.3	Too big
RTX 3060 12GB	12 GB	Q5_K_M	2.2	Too big
RTX 4060 8GB	8 GB	Q5_K_M	2.0	Too big

Estimates for single-stream decoding at Q5_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 InternLM 20B (20B params) fits at each quantization level on Mac Studio M2 Ultra 128GB (92.2 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	7.8 GB	Low	C47
Q3_K_S	3	9.8 GB	Low	C47
NVFP4	4	11.2 GB	Medium	C48
Q4_K_M	4	12.2 GB	Medium	C48
Q5_K_M	5	14.4 GB	High	C48
Q6_K	6	16.4 GB	High	C48
Q8_0	8	21.4 GB	Very High	C49
F16Best for your GPU	16	41.0 GB	Maximum	C53

Get started

Copy-paste commands to run InternLM 20B on your machine.

Run

docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
  --hf-repo "internlm/internlm2_5-20b-chat" \
  --hf-file "internlm2_5-20b-chat-Q5_K_M.gguf" \
  -c 4096 -ngl 99

Opções de upgrade

Hardware que roda bem InternLM 20B

RTX PRO 6000 Blackwell Workstation Edition 96GBOpção econômica

1792 GB/s (+992)

Raises estimated decode speed by about 224%.106.6 tok/s decodificação

Raises estimated decode speed by about 224%.

~$9,999 MSRP

RTX PRO 6000 Blackwell Server Edition 96GBMelhor custo-benefício

1597 GB/s (+797)

Raises estimated decode speed by about 189%.95 tok/s decodificação

Raises estimated decode speed by about 189%.

~$9,999 MSRP

Frequently asked questions

Can Mac Studio M2 Ultra 128GB run InternLM 20B?

Yes, Mac Studio M2 Ultra 128GB can run InternLM 20B with a B grade (Runs well). Expected decode speed: 32.9 tok/s.

How much VRAM does InternLM 20B need?

InternLM 20B (20B parameters) requires approximately 47.4 GB of memory with Q5_K_M quantization.

What is the best quantization for InternLM 20B?

The recommended quantization for InternLM 20B is Q5_K_M, which balances quality and memory efficiency.

What speed will InternLM 20B run at on Mac Studio M2 Ultra 128GB?

On Mac Studio M2 Ultra 128GB, InternLM 20B achieves approximately 32.9 tokens per second decode speed with a time-to-first-token of 5890ms using Q5_K_M quantization.

Can Mac Studio M2 Ultra 128GB run InternLM 20B for coding?

For coding workloads, InternLM 20B on Mac Studio M2 Ultra 128GB receives a B grade with 32.9 tok/s and 8K context.

What context window can InternLM 20B use on Mac Studio M2 Ultra 128GB?

On Mac Studio M2 Ultra 128GB, InternLM 20B can safely use up to 8K tokens of context. The model's official context limit is 8K, but available memory constrains the safe maximum.

Is unified memory on Mac Studio M2 Ultra 128GB as fast as VRAM for InternLM 20B?

Not always. Mac Studio M2 Ultra 128GB can often fit larger models thanks to unified memory, but a discrete GPU with dedicated high-bandwidth VRAM may still decode faster once the model fits. For this combination, the important distinction is capacity versus sustained throughput.

See all results for Mac Studio M2 Ultra 128GB See all hardware for InternLM 20B

Embed this result▼

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

<iframe src="https://willitrunai.com/embed/internlm-20b-on-m2-ultra-128gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: