K2-Horizon (375B-A23B) on AMD Instinct MI300X

Can the AMD Instinct MI300X run K2-Horizon (375B-A23B)?

Yes, it runs well (tight fit)

Runs well (tight fit). At Q2_K it needs about 164 GB of 192 GB usable, with an estimated speed of 178 tok/s.

Best quant: Q2_K
Generation speed: 178 tok/s
Prompt processing: 4,405 tok/s
First token (1K prompt): 0.3 s
Largest safe context: 117K

Best app to run it on the AMD Instinct MI300X

On Linux

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 bartowski/K2-Horizon-375B-A23B-GGUF:Q2_K

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

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 bartowski/K2-Horizon-375B-A23B-GGUF:Q2_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? K2-Horizon (375B-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 K2-Horizon (375B-A23B)

Recommended for the AMD Instinct MI300X: Q2_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 Q2_K files (opens in a new tab)

huggingface.co/bartowski/K2-Horizon-375B-A23B-GGUF (opens in a new tab)

ollama run hf.co/bartowski/K2-Horizon-375B-A23B-GGUF

Original weights · Hugging Face

Full-precision files from Institute of Foundation Models, for fine-tuning, vLLM, Transformers.

huggingface.co/IFM/K2-Horizon-375B-A23B (opens in a new tab)

No sign-in needed.

Official repo · GitHub

Reference code, inference scripts and docs from Institute of Foundation Models. The weights themselves download from Hugging Face.

github.com/ifm-ai/xllm (opens in a new tab)

IFM’s open training code (xLLM) for the K2-Horizon family.

git clone https://github.com/ifm-ai/xllm

Links checked 1 day ago against Hugging Face, Ollama and GitHub.

How it’s rated

Quality 66Good quality (B) · 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking.: How capable the model itself is at Q2_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 welltight fit: How well it runs on the AMD Instinct MI300X: 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.	29.8 / 45
Memory fitFull marks when everything sits in GPU memory; fewer when part spills to system RAM.	24.0 / 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	83.8 / 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 MI300X

Quant	Memory needed	Available	Speed	Runs	Quality	Download
`Q2_K`	164 GB	192 GB	178 tok/s	Runs welltight fit	66Good quality (B) · 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking.	GGUF · folder (opens in a new tab)
`Q3_K_M`	198 GB	192 GB	74.7 tok/s	Runs welloffloads to RAM	79Great quality (A) · 79 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking.	GGUF · folder (opens in a new tab)
`Q4_K_M`	244 GB	192 GB	—	Won't run	89Excellent quality (A+) · 89 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 12/100 (quality, fit, speed and memory use) sets the ranking.	GGUF · folder (opens in a new tab)
`Q5_K_M`	281 GB	192 GB	—	Won't run	91Excellent quality (A+) · 91 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 12/100 (quality, fit, speed and memory use) sets the ranking.	GGUF · folder (opens in a new tab)
`Q6_K`	323 GB	192 GB	—	Won't run	93Excellent quality (A+) · 93 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 12/100 (quality, fit, speed and memory use) sets the ranking.	GGUF · folder (opens in a new tab)
`Q8_0`	414 GB	192 GB	—	Won't run	94Excellent quality (A+) · 94 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 12/100 (quality, fit, speed and memory use) sets the ranking.	GGUF · folder (opens in a new tab)
`F16`	771 GB	192 GB	—	Won't run	94Excellent quality (A+) · 94 of 100: how capable the model is at F16, whatever it runs on. Match score 12/100 (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

89Excellent quality (A+) · 89 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking. Apple M3 Ultra (512 GB) Q4_K_M · 33.4 tok/s · Runs great
66Good quality (B) · 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. Apple M3 Ultra (256 GB) Q2_K · 51.8 tok/s · Runs well (tight fit)
91Excellent quality (A+) · 91 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking. Apple M5 Ultra (512 GB) Q5_K_M · 39.3 tok/s · Runs great
91Excellent quality (A+) · 91 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. AMD Instinct MI350X Q5_K_M · 147 tok/s · Runs well (tight fit)
89Excellent quality (A+) · 89 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking. AMD Instinct MI325X Q4_K_M · 143 tok/s · Runs well (tight fit)
89Excellent quality (A+) · 89 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip Q4_K_M · 167 tok/s · Runs well (tight fit)

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.