IFM/K2-Horizon-MoVA-36B-A4B
text-generation model by IFM
Older snapshot: check the original source for current information.
From the publisher
Model card & documentation
Source previewRead the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.
K2-Horizon-MoVA-36B-A4B is the sparse member of the K2-Horizon family: a Mixture-of-Experts model with Mixture-of-Values attention (MoVA) that stores 36B parameters and runs 4B per token. We have released the checkpoints, along with the data and the training code. Open-weight modelsK2-Horizon-MoVA-36B-A4BNemotron 3 UltraNemotron 3 SuperG9v3-39A5BQwen3.6-35B-A3BMuse Glimmer-30BGemma 4 31B-it# Params36B550B120B39B35B30B31B# Activated params4B55B12B5B3B30B31BArchitectureMoEMoEMoEMoEMoEDenseDenseAgentstau3-BankingAgentic tool use26.814.210.322.19.323.514.8CodingTerminal-Bench 2.1Agentic terminal use58.653.938.632.644.951.743.4SciCodeScientific coding38.939.936.034.035.843.643.4Scientific ReasoningHumanity's Last Exam (without tools)Expert-level reasoning25.228.420.817.522.222.023.6GPQA DiamondGraduate-level science QA80.886.780.080.584.183.585.7CritPtFrontier physics reasoning2.13.13.10.30.32.61.4GeneralAA-LCRLong-context reasoning66.371.060.362.066.780.068.3AA-Omniscience AccuracyFactual accuracy18.822.624.314.918.827.020.0AA-Omniscience Non-HallucinationNon-hallucination rate69.270.313.087.049.518.115.0 Scores in %. Bold marks the best score in each row. Sections follow the Artificial Analysis Intelligence Index categories. Baseline scores are from Artificial Analysis; Muse Glimmer-30B at high reasoning effort, all other open models in their reasoning mode. vLLM, recipe at recipes.vllm.ai/IFM: Use an exact branch name from the inventory with vLLM's --revision option. For example, --revision pretrain1100000 selects the final checkpoint of Pretraining, at step 1,100,000.…
Read the full model card ↗ · Preview checked 2026-10-30T11:09:34.842Z
Inside the original model card — Document outline
- K2-Horizon-MoVA-36B-A4B
- K2-Horizon-MoVA-36B-A4B Highlights
- Benchmark Results
- Quickstart
- Serving
- API Usage
- Transformers
- Training Overview
- Release Artifacts
- Artifact Index
- Checkpoint Inventory
- Best Practices
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Links from the model card
References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.
Documentation belongs to its respective authors. Reported project/model license: apache-2.0. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
IFM/K2-Horizon-MoVA-36B-A4B is a text-generation model repository published by IFM on Hugging Face. The source reports the transformers library.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- text-generation
- Library
- transformers
- Recent downloads (30 days)
- 9,445
- Cumulative likes
- 335
- Hugging Face trending score
- 60
- Reported safetensors parameters
- 37,444,792,020
- Architecture
- K2HorizonForCausalLM
- License
- apache-2.0
- Access
- Not gated by Hugging Face
- Created
- 2026-09-01T23:01:25.000Z
- Last modified
- 2026-09-15T21:57:31.000Z
Compatibility and lineage
Reported languages: en
Base-model lineage was not included.
Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.
Use and evaluate
Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.
No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.
Source and freshness
Source: Hugging Face Hub. Metadata observed 2026-09-18T06:35:08.044Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.