No official torrent yet — requesting it tells the flock what to host next. You cannot seed a catalog-only page.
Cataloged from IFM/K2-Horizon-375B-A23B-FP8
All models by IFM →Training Code - Evaluation Code - Pretraining Data - Midtraining Data
[!NOTE] This repository contains an FP8-quantized version of IFM/K2-Horizon-375B-A23B.
Only the routed-expert linear layers are quantized to FP8:
- Weights: static FP8, one scale per 128*128 block.
- Activations: dynamic FP8, one scale per 1*128 group along the input-channel dim.
All other linear layers (attention, shared experts, routers, the first 3 dense layers, and lm_head) are kept in BF16.
The FP8 model performs closely in line with the original BF16 model on our evaluations, while reducing memory footprint and enabling faster inference on FP8-capable hardware.
Serving note: the routed experts' intermediate size (1792) is not splittable into whole 128-wide quantization blocks at the usual tensor-parallel sizes (TP=4, TP=8), so expert parallelism is required.
K2-Horizon-375B-A23B is the flagship of the K2-Horizon family: a sparse Mixture-of-Experts model that stores 375B parameters and runs 23B per token, with a 512K context window. We have released the final checkpoint; intermediate checkpoints, along with the data and the training code, will be released.
[Image blocked: K2-Horizon-375B-A23B benchmark results against open MoE, dense, and closed models]
Scores in %, except GDPVal-AA, which is an Elo rating. Bold marks the best score in each row. The first four sections follow the Artificial Analysis Intelligence Index categories. Baseline scores are from Artificial Analysis where available, otherwise from the IFM evaluation harness. SWE-Atlas-QnA and SWE Bench Pro are run without internet access; BrowseComp uses the Discard-all@95k context setting from the DeepSeek-V3.2 technical report; WildClawBench and Apex-Agents use the English text-only subsets.
vLLM, recipe at recipes.vllm.ai/IFM:
SGLang, adapted from the BF16 recipe validated on 8× H200 in the SGLang K2 Horizon cookbook:
[!Tip] Recommended settings:
reasoning_effort="high",temperature=1.0,top_p=0.95, and at least 32,768 output tokens. Reasoning depth is selected per request throughchat_template_kwargs. Thinking is returned inreasoning_contentand the answer incontent.
Our model supports multiple tool calls formats, which can be changed with chat_template_kwargs. The supported values are json, xml, and xml_typed . The default is xml. Keep --tool-call-parser k2_horizon enabled to parse the selected format.
Validated with Transformers 4.57.6, PyTorch 2.13.0, Safetensors 0.8.0.
high. All reported results use high reasoning effort. Pass {"chat_template_kwargs": {"reasoning_effort": "high"}} on every request.temperature=1.0, top_p=0.95.--dtype flag is needed. Full recipes for every K2-Horizon size, with measured H200 latency and throughput, are in the SGLang cookbook and the vLLM recipe.k2_horizon reasoning parser for chat, and add the k2_horizon tool-call parser for agent use. Leave both off for plain completion-style generation.vllm serve IFM/K2-Horizon-375B-A23B-FP8 \ --revision main \ --model-impl transformers \ --tensor-parallel-size 8 \ --enable-expert-parallel \ --trust-remote-code \ --max-model-len 131072 \ --reasoning-parser k2_horizon \ --tool-call-parser k2_horizon \ --enable-auto-tool-choicepython3 -m sglang.launch_server \ --model-path IFM/K2-Horizon-375B-A23B-FP8 \ --revision main \ --tp 8 \ --ep 8 \ --attention-backend fa3 \ --model-loader-extra-config '{"enable_multithread_load":false}' \ --reasoning-parser k2_horizon \ --tool-call-parser k2_horizon \ --host 0.0.0.0 --port 30000from openai import OpenAI
client = OpenAI(base_url="http://localhost:30000/v1", api_key="EMPTY")response = client.chat.completions.create( model="IFM/K2-Horizon-375B-A23B-FP8", messages=[{"role": "user", "content": "Explain the result step by step."}], temperature=1.0, top_p=0.95, max_tokens=32768, extra_body={"chat_template_kwargs": {"reasoning_effort": "high", "tool_call_format": "xml"}},)message = response.choices[0].messageprint("Reasoning:", getattr(message, "reasoning_content", None))print("Answer:", message.content)from transformers import AutoModelForCausalLM, AutoTokenizer
model_id = "IFM/K2-Horizon-375B-A23B-FP8"tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)model = AutoModelForCausalLM.from_pretrained( model_id, device_map="auto", dtype="auto", low_cpu_mem_usage=True, trust_remote_code=True)
inputs = tokenizer("Explain why long-context evaluation is difficult.", return_tensors="pt").to(model.device)inputs.pop("token_type_ids", None)outputs = model.generate(**inputs, max_new_tokens=32768, temperature=1.0, top_p=0.95, do_sample=True)print(tokenizer.decode(outputs[0], skip_special_tokens=True))@misc{k2horizon2026, title = {Introducing K2 Horizon: Frontier Performance, Radically Open}, author = {{IFM Team}}, year = {2026}, url = {https://ifm.ai/blog/k2/},}