on

Can the NVIDIA L40 run Phi-3.5 MoE (42B-A6.6B)?

55Good quality (B) · 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.

Yes, it runs great

Runs great. At Q6_K it needs about 38.4 GB of 48.0 GB usable, with an estimated speed of 79.5 tok/s.

Best quant
Q6_K
Generation speed
79.5 tok/s
Prompt processing
9,873 tok/s
First token (1K prompt)
0.3 s
Largest safe context
79K

Best app to run it on the NVIDIA L40

On Linux

LM Studio

Desktop app · runs on llama.cpp, MLX

  • Point-and-click: search, download and chat in one window.

Open Phi-3.5 MoE (42B-A6.6B) in LM Studio

  • llama.cpp Usually the first to support new model architectures.
  • Ollama Runs the Hugging Face GGUF with ollama run hf.co/…

On Windows

LM Studio

Desktop app · runs on llama.cpp, MLX

  • Point-and-click: search, download and chat in one window.

Open Phi-3.5 MoE (42B-A6.6B) in LM Studio

  • llama.cpp Usually the first to support new model architectures.
  • Ollama Runs the Hugging Face GGUF with ollama run hf.co/…

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 Phi-3.5 MoE (42B-A6.6B)

Recommended for the NVIDIA L40: Q6_K. 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 Q6_K file (opens in a new tab)

huggingface.co/lmstudio-community/Phi-3.5-MoE-instruct-GGUF (opens in a new tab)

ollama run hf.co/lmstudio-community/Phi-3.5-MoE-instruct-GGUF:Q6_K

Original weights · Hugging Face

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

huggingface.co/microsoft/Phi-3.5-MoE-instruct (opens in a new tab)

No sign-in needed.

Official repo · GitHub

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

github.com/microsoft/PhiCookBook (opens in a new tab)

Microsoft’s official Phi cookbook: setup, inference and fine-tuning guides.

git clone https://github.com/microsoft/PhiCookBook

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

How it’s rated

Quality 55Good quality (B) · 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.
How capable the model itself is at Q6_K, 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 L40: 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.24.9 / 45
Memory fitFull marks when everything sits in GPU 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.15.0 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.15.0 / 15
Match score79.9 / 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 L40

Quant Memory needed Available Speed Runs Quality Download
Q2_K 20.8 GB 48.0 GB 145 tok/s Runs great 39Basic quality (D) · 39 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. GGUF files (opens in a new tab)
Q3_K_M 24.6 GB 48.0 GB 124 tok/s Runs great 47Fair quality (C) · 47 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. GGUF (Q3_K_L) (opens in a new tab)
Q4_K_M 29.6 GB 48.0 GB 104 tok/s Runs great 53Fair quality (C) · 53 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. GGUF (opens in a new tab)
Q5_K_M 33.8 GB 48.0 GB 90.6 tok/s Runs great 54Fair quality (C) · 54 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. GGUF files (opens in a new tab)
Q6_K 38.4 GB 48.0 GB 79.5 tok/s Runs great 55Good quality (B) · 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. GGUF (opens in a new tab)
Q8_0 48.5 GB 48.0 GB 57.0 tok/s Runs welloffloads to RAM 56Good quality (B) · 56 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. GGUF (opens in a new tab)
F16 87.8 GB 48.0 GB — Won't run 56Good quality (B) · 56 of 100: how capable the model is at F16, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. 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 Phi-3.5 MoE (42B-A6.6B) need on the NVIDIA L40 48GB?

At Q6_K it needs about 38.4 GB. The NVIDIA L40 48GB has 48.0 GB usable.

How fast is Phi-3.5 MoE (42B-A6.6B) on the NVIDIA L40 48GB?

About 79.5 tok/s at Q6_K, an estimate from the hardware’s memory bandwidth and the size of the model.

Which quantization of Phi-3.5 MoE (42B-A6.6B) should I use on the NVIDIA L40 48GB?

Q6_K. 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?