K-EXAONE 236B-A23B on AMD Instinct MI350X

Can the AMD Instinct MI350X run K-EXAONE 236B-A23B?

Yes, it runs great

Runs great. At Q6_K it needs about 210 GB of 288 GB usable, with an estimated speed of 133 tok/s.

Best quant: Q6_K
Generation speed: 133 tok/s
Prompt processing: 3,693 tok/s
First token (1K prompt): 0.3 s
Largest safe context: 256K

Best app to run it on the AMD Instinct MI350X

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 LGAI-EXAONE/K-EXAONE-236B-A23B

Lemonade AMD’s own app: llama.cpp tuned for Radeon on Vulkan and ROCm.
LM Studio Point-and-click: search, download and chat in one window.

On Windows

Lemonade

Desktop app · runs on llama.cpp, ONNX Runtime GenAI (Ryzen AI NPU)

AMD’s own app: llama.cpp tuned for Radeon on Vulkan and ROCm.

lemonade pull LGAI-EXAONE/K-EXAONE-236B-A23B-GGUF:Q6_K

LM Studio Point-and-click: search, download and chat in one window.
llama.cpp Usually the first to support new model architectures.

Want an agent? K-EXAONE 236B-A23B 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 K-EXAONE 236B-A23B

Recommended for the AMD Instinct MI350X: 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/LGAI-EXAONE/K-EXAONE-236B-A23B-GGUF (opens in a new tab)

ollama run hf.co/LGAI-EXAONE/K-EXAONE-236B-A23B-GGUF:Q6_K

Original weights · Hugging Face

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

huggingface.co/LGAI-EXAONE/K-EXAONE-236B-A23B (opens in a new tab)

No sign-in needed.

Official repo · GitHub

Reference code, inference scripts and docs from LG AI Research. The weights themselves download from Hugging Face.

github.com/LG-AI-EXAONE/K-EXAONE (opens in a new tab)

git clone https://github.com/LG-AI-EXAONE/K-EXAONE

Links checked 2 days ago against Hugging Face, Ollama and GitHub.

How it’s rated

Quality 87Excellent quality (A+) · 87 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 94/100 (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 AMD Instinct MI350X: 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.	39.2 / 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 score	94.2 / 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 AMD Instinct MI350X

Quant	Memory needed	Available	Speed	Runs	Quality	Download
`Q2_K`	110 GB	288 GB	209 tok/s	Runs great	61Good quality (B) · 61 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`Q3_K_M`	132 GB	288 GB	187 tok/s	Runs great	73Great quality (A) · 73 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`Q4_K_M`	160 GB	288 GB	164 tok/s	Runs great	83Great quality (A) · 83 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 90/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q5_K_M`	184 GB	288 GB	147 tok/s	Runs great	85Excellent quality (A+) · 85 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q6_K`	210 GB	288 GB	133 tok/s	Runs great	87Excellent quality (A+) · 87 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q8_0`	267 GB	288 GB	109 tok/s	Runs welltight fit	88Excellent quality (A+) · 88 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.	GGUF (opens in a new tab)
`F16`	490 GB	288 GB	—	Won't run	88Excellent quality (A+) · 88 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.	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

83Great quality (A) · 83 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. Apple M3 Ultra (256 GB) Q4_K_M · 33.4 tok/s · Runs great
73Great quality (A) · 73 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. Apple M2 Ultra (192 GB) Q3_K_M · 41.2 tok/s · Runs well (tight fit)
83Great quality (A) · 83 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. Apple M3 Ultra (512 GB) Q4_K_M · 33.4 tok/s · Runs great
61Good quality (B) · 61 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GB10 (128 GB) Q2_K · 22.3 tok/s · Runs well (tight fit)
61Good quality (B) · 61 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. NVIDIA RTX PRO 6000 Blackwell Q2_K · 56.8 tok/s · Runs well (offloads to RAM)
61Good quality (B) · 61 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. Apple M4 Max (128 GB) Q2_K · 31.0 tok/s · Runs well (over GPU memory limit)

Quick answers

How much memory does K-EXAONE 236B-A23B need on the AMD Instinct MI350X 288GB?

At Q6_K it needs about 210 GB. The AMD Instinct MI350X 288GB has 288 GB usable.

How fast is K-EXAONE 236B-A23B on the AMD Instinct MI350X 288GB?

About 133 tok/s at Q6_K, an estimate from the hardware’s memory bandwidth and the size of the model.

Which quantization of K-EXAONE 236B-A23B should I use on the AMD Instinct MI350X 288GB?

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.