---

pipeline_tag: text-generation

base_model:

• IFM/K2-Horizon-3.7B

library_name: llama-cpp

tags:

• k2-horizon
• 3.7b
• dense
• open-weights
• ifm
• gguf
• nvfp4

language:

• en

license: apache-2.0

license_name: apache-license-2.0

license_link: https://apache.org/licenses/LICENSE-2.0

---

K2-Horizon-3.7B NVFP4 — GGUF

NVFP4 GGUF conversions derived from the K2-Horizon-3.7B model family. This repository provides NVFP4 (NVIDIA 4-bit floating-point) GGUF files, making the model usable with llama.cpp and other GGUF-compatible inference engines.

Model Overview

K2-Horizon-3.7B is the small dense member of the K2-Horizon family: a 3.7B-core decoder-only model with a 512K context window.

Model Architecture

| Property | Value |

| :----------------------- | :-------------------- |

| Architecture | K2-Horizon-3.7B |

| Parameters | 3.7B |

| Context Length | 512K tokens (524,288) |

| Supported Modalities | Text |

| Type | Dense |

Benchmark Results

| Benchmark | K2-Horizon-3.7B | Qwen3.5-4B | G9v3-3B | Granite 4.2-3B | Nemotron 3 Nano-4B |

| :----------------------- | :-----------: | :--------: | :-----: | :-------------: | :----------------: |

| Math | | | | | |

| HMMT Feb 2026 | 70.5 | 61.6 | 34.1 | 57.2 | 34.7 |

| Coding | | | | | |

| SWE-bench Verified | 68.6 | 41.2 | 16.4 | 32.2 | 1.8 |

| Scientific Reasoning | | | | | |

| GPQA Diamond | 65.4 | 77.1 | 43.8 | 55.9 | 51.3 |

| HLE | 12.9 | 9.9 | 4.5 | 6.6 | 4.9 |

| SciCode | 25.9 | 16.1 | 17.7 | 24.9 | 16.4 |

| Agents | | | | | |

| Terminal-Bench 2.1 | 25.1 | 25.8 | 6.0 | 13.9 | 3.7 |

| tau3-Banking | 17.7 | 6.8 | — | 5.6 | — |

| BFCL v4 | 50.9 | 55.7 | 47.9 | 50.8 | 36.8 |

Scores in %. Bold marks the best score in each row. Benchmark results are sourced from the original K2-Horizon-3.7B model description.

GGUF Files

| File | Format | Description |

| :------------------------- | :----- | :-------------------- |

| K2-Horizon-3.7B-nvfp4.gguf | NVFP4 | NVFP4 quantized model |

Usage

llama.cpp (CLI)

# Run inference
./llama-cli \
  -m K2-Horizon-3.7B-nvfp4.gguf \
  --temp 1.0 --top-p 0.95

llama-server (OpenAI-compatible API)

# Start the server
./llama-server \
  -m K2-Horizon-3.7B-nvfp4.gguf \
  --port 8080

Key Features

• Strong small-model baseline. A dense model evaluated on the same agentic, coding, and reasoning benchmarks as the rest of the family.
• 512K context. Native 524,288-token context from the midtraining stages onward.
• NVFP4 Quantization. Optimized for NVIDIA Blackwell and compatible hardware.

Citation

@misc{k2horizon2026,
  title  = {Introducing K2 Horizon: Frontier Performance, Radically Open},
  author = {{IFM Team}},
  year   = {2026},
  url    = {https://ifm.ai/blog/k2/},
}