Qwen3-Next 80B-A3B Thinking on NVIDIA GeForce RTX 4090

Can the NVIDIA GeForce RTX 4090 run Qwen3-Next 80B-A3B Thinking?

Yes, but slowly (heavy RAM offload)

Slow (heavy RAM offload). At Q3_K_M it needs about 41.8 GB of 24.0 GB usable, with an estimated speed of 58.9 tok/s.

Best quant: Q3_K_M
Generation speed: 58.9 tok/s
Prompt processing: 13,608 tok/s
First token (1K prompt): 0.3 s
Largest safe context: —

Best app to run it on the NVIDIA GeForce RTX 4090

On Windows

LM Studio

Desktop app · runs on llama.cpp, MLX

Splits the model between GPU memory and system RAM, which this setup needs.
Point-and-click: search, download and chat in one window.

Open Qwen3-Next 80B-A3B Thinking in LM Studio

llama.cpp Splits the model between GPU memory and system RAM, which this setup needs.
Ollama Splits the model between GPU memory and system RAM, which this setup needs.

On Linux

LM Studio

Desktop app · runs on llama.cpp, MLX

Splits the model between GPU memory and system RAM, which this setup needs.
Point-and-click: search, download and chat in one window.

Open Qwen3-Next 80B-A3B Thinking in LM Studio

llama.cpp Splits the model between GPU memory and system RAM, which this setup needs.
Ollama Splits the model between GPU memory and system RAM, which this setup needs.

Want an agent? Qwen3-Next 80B-A3B Thinking supports tool use, so you can pair a llama.cpp or MLX LM server with one of these. Apps that run on this computer come first.

Coding agents

Personal assistants

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 Qwen3-Next 80B-A3B Thinking

Quantized files · GGUF

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

huggingface.co/Qwen/Qwen3-Next-80B-A3B-Thinking-GGUF (opens in a new tab)

ollama run hf.co/Qwen/Qwen3-Next-80B-A3B-Thinking-GGUF

Original weights · Hugging Face

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

huggingface.co/Qwen/Qwen3-Next-80B-A3B-Thinking (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 72Great quality (A) · 72 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.: How capable the model itself is at Q3_K_M, 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 Slowheavy RAM offload: How well it runs on the NVIDIA GeForce RTX 4090: 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.	32.5 / 45
Memory fitFull marks when everything sits in GPU memory; fewer when part spills to system RAM.	4.0 / 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 score	66.5 / 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 GeForce RTX 4090

Quant	Memory needed	Available	Speed	Runs	Quality	Download
`Q2_K`	34.5 GB	24.0 GB	88.1 tok/s	Slowheavy RAM offload	60Good quality (B) · 60 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`Q3_K_M`	41.8 GB	24.0 GB	58.9 tok/s	Slowheavy RAM offload	72Great quality (A) · 72 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`	51.6 GB	24.0 GB	—	Won't run	81Great quality (A) · 81 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 11/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q5_K_M`	59.7 GB	24.0 GB	—	Won't run	83Great quality (A) · 83 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 11/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q6_K`	68.7 GB	24.0 GB	—	Won't run	85Excellent quality (A+) · 85 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 11/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q8_0`	88.2 GB	24.0 GB	—	Won't run	86Excellent quality (A+) · 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 11/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`F16`	165 GB	24.0 GB	—	Won't run	86Excellent quality (A+) · 86 of 100: how capable the model is at F16, whatever it runs on. Match score 11/100 (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

86Excellent quality (A+) · 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. Apple M2 Ultra (192 GB) Q8_0 · 92.8 tok/s · Runs great
86Excellent quality (A+) · 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GB10 (128 GB) Q8_0 · 46.8 tok/s · Runs great
85Excellent quality (A+) · 85 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. AMD Ryzen AI Max+ 395 (128 GB) Q6_K · 51.1 tok/s · Runs great
85Excellent quality (A+) · 85 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. Apple M4 Max (128 GB) Q6_K · 90.1 tok/s · Runs great
85Excellent quality (A+) · 85 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA RTX PRO 6000 Blackwell Q6_K · 179 tok/s · Runs great
85Excellent quality (A+) · 85 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA RTX Spark N1X (6144-core GPU) (128 GB) Q6_K · 65.1 tok/s · Runs great

Quick answers

How much memory does Qwen3-Next 80B-A3B Thinking need on the RTX 4090 24GB?

At Q3_K_M it needs about 41.8 GB. The RTX 4090 24GB has 24.0 GB usable.

How fast is Qwen3-Next 80B-A3B Thinking on the RTX 4090 24GB?

About 58.9 tok/s at Q3_K_M, an estimate from the hardware’s memory bandwidth and the size of the model.

Which quantization of Qwen3-Next 80B-A3B Thinking should I use on the RTX 4090 24GB?

Q3_K_M. 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.