Check my GPU

on

Can the Intel Core Ultra 200H (Arc 140T) (96 GB) run Qwen3.8 27B?

83 ptsGreat (A) · 83 of 100 points: Model quality 42.6/45, Memory fit 25.0/25, Speed 3.7/15, Memory use 11.9/15

Yes, it runs well

Runs well. At Q6_K it needs about 28.3 GB of 54.7 GB usable, with an estimated speed of 2.4 tok/s.

Best quant
Q6_K
Generation speed
2.4 tok/s
Prompt processing
182 tok/s
First token (1K prompt)
5.6 s
Largest safe context
256K

Download Qwen3.8 27B

Recommended for the Intel Core Ultra 200H (Arc 140T) (96 GB): Q6_K. Every other quantization has its own download link in the table below.

Quickest · Ollama

One command downloads a ready-to-run version and starts a chat.

ollama run qwen3.8:27b

Get Ollama at ollama.com/download (opens in a new tab) · model page (opens in a new tab)

Quantized files · GGUF

Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.

Download the Q6_K file (opens in a new tab)

huggingface.co/unsloth/Qwen3.8-27B-GGUF (opens in a new tab)

ollama run hf.co/unsloth/Qwen3.8-27B-GGUF:Q6_K

Original weights · Hugging Face

Full-precision files from Alibaba, for fine-tuning, vLLM, Transformers.

huggingface.co/Qwen/Qwen3.8-27B (opens in a new tab)

No sign-in needed.

Official repo · GitHub

Reference code, inference scripts and docs from Alibaba. The weights themselves download from Hugging Face.

github.com/QwenLM/Qwen3.8 (opens in a new tab)

git clone https://github.com/QwenLM/Qwen3.8

Qwen’s current repo: Qwen3.5 and Qwen3.6 redirect here.

Links checked 5 days ago against Hugging Face, Ollama and GitHub.

How the A grade is scored

Every result gets up to 100 points from four parts. The points decide the letter grade.

Model qualityEditorial quality tier of the model, scaled by the quantization or precision used.42.6 / 45
Memory fitFull marks when everything sits in GPU or unified memory; fewer when part spills to system RAM.25.0 / 25
SpeedGeneration speed against a 40 tokens/s target for LLMs, or render time against a target for image and video.3.7 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.11.9 / 15
Total83.2 / 10083 ptsGreat (A) · 83 of 100 points: Model quality 42.6/45, Memory fit 25.0/25, Speed 3.7/15, Memory use 11.9/15

Grade scale: S 85+ · A 72+ · B 58+ · C 42+ · D 25+ · F under 25 points.

Every quantization on the Intel Core Ultra 200H (Arc 140T) (96 GB)

Quant Memory needed Available Speed Fit Grade · points Download
Q2_K 16.6 GB 54.7 GB 5.7 tok/s Runs well 67 ptsGood (B) · 67 of 100 points: Model quality 30.1/45, Memory fit 25.0/25, Speed 5.7/15, Memory use 7.0/15 GGUF (UD-Q2_K_XL) (opens in a new tab)
Q3_K_M 19.1 GB 54.7 GB 4.5 tok/s Runs well 74 ptsGreat (A) · 74 of 100 points: Model quality 36.1/45, Memory fit 25.0/25, Speed 5.0/15, Memory use 8.1/15 GGUF (UD-Q3_K_XL) (opens in a new tab)
Q4_K_M 22.5 GB 54.7 GB 3.5 tok/s Runs well 79 ptsGreat (A) · 79 of 100 points: Model quality 40.9/45, Memory fit 25.0/25, Speed 4.4/15, Memory use 9.5/15 GGUF (opens in a new tab)
Q5_K_M 25.2 GB 54.7 GB 2.9 tok/s Runs well 81 ptsGreat (A) · 81 of 100 points: Model quality 41.7/45, Memory fit 25.0/25, Speed 4.0/15, Memory use 10.6/15 GGUF (opens in a new tab)
Q6_K 28.3 GB 54.7 GB 2.4 tok/s Runs well 83 ptsGreat (A) · 83 of 100 points: Model quality 42.6/45, Memory fit 25.0/25, Speed 3.7/15, Memory use 11.9/15 GGUF (opens in a new tab)
Q8_0 35.0 GB 54.7 GB 1.8 tok/s Runs well 85 ptsExcellent (S) · 85 of 100 points: Model quality 43.0/45, Memory fit 25.0/25, Speed 3.1/15, Memory use 14.7/15 GGUF (opens in a new tab)
F16 61.1 GB 54.7 GB 0.7 tok/s Heavy offload 63 ptsGood (B) · 63 of 100 points: Model quality 43.0/45, Memory fit 4.0/25, Speed 2.0/15, Memory use 15.0/15 GGUF · folder (opens in a new tab) Original weights (opens in a new tab)

Memory figures assume 8K of context. Longer conversations or documents grow the KV cache.

Other hardware that runs it

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.