on

Can the NVIDIA GB10 (128 GB) run Kimi-VL A3B Thinking 2506?

60Good quality (B) · 60 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.

Yes, it runs great

Runs great. At Q8_0 it needs about 27.5 GB of 115 GB usable, with an estimated speed of 46.8 tok/s.

Best quant
Q8_0
Generation speed
46.8 tok/s
Prompt processing
7,440 tok/s
First token (1K prompt)
0.3 s
Largest safe context
128K

Best app to run it on the NVIDIA GB10 (128 GB)

On Linux

vLLM

Inference server · runs on vLLM

  • The full model fits in GPU memory at 8-bit, so it can batch many requests: best for serving a team or an app.
vllm serve moonshotai/Kimi-VL-A3B-Thinking-2506
  • LM Studio Point-and-click: search, download and chat in one window.
  • llama.cpp Usually the first to support new model architectures.

How we choose: the files published for this model, your operating system and GPU, whether it fits fully in memory, and published speed tests (MLX vs llama.cpp on Apple silicon, vLLM vs llama.cpp vs Ollama on NVIDIA). Details.

Download Kimi-VL A3B Thinking 2506

Recommended for the NVIDIA GB10 (128 GB): Q8_0. Every other quantization has its own download link in the table below.

Quantized files · GGUF

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

Download the Q8_0 file (opens in a new tab)

huggingface.co/ggml-org/Kimi-VL-A3B-Thinking-2506-GGUF (opens in a new tab)

ollama run hf.co/ggml-org/Kimi-VL-A3B-Thinking-2506-GGUF:Q8_0

Original weights · Hugging Face

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

huggingface.co/moonshotai/Kimi-VL-A3B-Thinking-2506 (opens in a new tab)

No sign-in needed.

Links checked 3 days ago against Hugging Face, Ollama and GitHub.

How it’s rated

Quality 60Good quality (B) · 60 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.
How capable the model itself is at Q8_0, from its editorial quality tier scaled by the quantization or precision. It’s the same on any hardware that runs this variant. Scale: A+ 85+ · A 70+ · B 55+ · C 40+ · D under 40.
Runs Runs great
How well it runs on the NVIDIA GB10 (128 GB): speed, capped by how well it fits in memory. Runs great: 30+ tokens/s (or near the image, video or decision speed target) · Runs well: 10+ tokens/s · Slow: 3+ tokens/s · Very slow: under 3 · Won’t run: doesn’t fit even with offload. A tight fit or RAM offload caps it at Runs well; heavy offload caps it at Slow.

Match score

To order lists, each result also gets a match score out of 100 from four parts. It favors the most capable model that still runs well.

Model qualityEditorial quality tier of the model, scaled by the quantization or precision used.27.0 / 45
Memory fitFull marks when the model fits in the share of unified memory the GPU can use; fewer when it goes past that limit.25.0 / 25
SpeedGeneration speed against a 40 tokens/s target for LLMs, or render time against a target for image and video.15.0 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.5.5 / 15
Match score72.5 / 100

The match score orders results; it is not shown as a grade. Quality and Runs are the two ratings in the tables.

Every quantization on the NVIDIA GB10 (128 GB)

Quant Memory needed Available Speed Runs Quality Download
Q2_K 16.6 GB 115 GB 117 tok/s Runs great 42Fair quality (C) · 42 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. GGUF files (opens in a new tab)
Q3_K_M 18.1 GB 115 GB 98.4 tok/s Runs great 50Fair quality (C) · 50 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. GGUF files (opens in a new tab)
Q4_K_M 20.1 GB 115 GB 81.2 tok/s Runs great 57Good quality (B) · 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. GGUF (opens in a new tab)
Q5_K_M 21.7 GB 115 GB 69.8 tok/s Runs great 58Good quality (B) · 58 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. GGUF files (opens in a new tab)
Q6_K 23.5 GB 115 GB 60.6 tok/s Runs great 59Good quality (B) · 59 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. GGUF files (opens in a new tab)
Q8_0 27.5 GB 115 GB 46.8 tok/s Runs great 60Good quality (B) · 60 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. GGUF (opens in a new tab)
F16 42.9 GB 115 GB 23.5 tok/s Runs well 60Good quality (B) · 60 of 100: how capable the model is at F16, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. GGUF (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

Quick answers

How much memory does Kimi-VL A3B Thinking 2506 need on the NVIDIA GB10 128GB?

At Q8_0 it needs about 27.5 GB. The NVIDIA GB10 128GB has 115 GB usable.

How fast is Kimi-VL A3B Thinking 2506 on the NVIDIA GB10 128GB?

About 46.8 tok/s at Q8_0, an estimate from the hardware’s memory bandwidth and the size of the model.

Which quantization of Kimi-VL A3B Thinking 2506 should I use on the NVIDIA GB10 128GB?

Q8_0. Of the versions that fit, it is the one that runs best; when two run equally well, the check prefers the higher-rated one.

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.

Was this page helpful?