K2 Horizon 0.9B

Institute of Foundation Models

K2 Horizon 0.9B is the compact dense member of IFM's (MBZUAI) fully open K2 Horizon family: a 0.9B-class decoder-only reasoning model (28 layers, smaller 64K vocabulary; about 1.08B parameters in the released safetensors) pretrained on 5T tokens and post-trained with multi-teacher distillation from math/code, STEM, and instruction-following experts followed by on-policy distillation. Supports English and Chinese and a 131,072-token (128K) context window via YaRN RoPE scaling. Text in, text out, with per-request reasoning effort (high recommended) and JSON/XML tool-call formats. Released 2026-09-03 under Apache 2.0 with BF16 weights plus a GGUF build, an optional Uno discrete-diffusion LoRA adapter for faster decoding, intermediate checkpoints, and training code. No priced provider is cataloged yet.

Key Specifications

Parameters
900.0M
Context
-
Release Date
September 3, 2026
Average Score
44.6%

Timeline

Key dates in the model's history
Announcement
September 3, 2026
Last Update
October 31, 2026
Today
November 20, 2026

Technical Specifications

Parameters
900.0M
Training Tokens
5.0T tokens
Knowledge Cutoff
-
Family
-
Capabilities
MultimodalZeroEval

Benchmark Results

Model performance metrics across various tests and benchmarks

Reasoning

Logical reasoning and analysis
GPQA
IFM K2-Horizon-0.9B Hugging Face model card, final checkpoint; reasoning_effort=high, temperature=0.6, top_p=0.95; GPQA Diamond • Self-reported
27.3%

Other Tests

Specialized benchmarks
AIME 2025
IFM K2-Horizon-0.9B Hugging Face model card, final checkpoint; reasoning_effort=high, temperature=0.6, top_p=0.95 • Self-reported
41.7%
AIME 2026
IFM K2-Horizon-0.9B Hugging Face model card, final checkpoint; reasoning_effort=high, temperature=0.6, top_p=0.95 • Self-reported
48.5%
BFCL-V4
IFM K2-Horizon-0.9B Hugging Face model card, final checkpoint; reasoning_effort=high, temperature=0.6, top_p=0.95 • Self-reported
28.0%
HMMT Feb 26
IFM K2-Horizon-0.9B Hugging Face model card, final checkpoint; reasoning_effort=high, temperature=0.6, top_p=0.95; HMMT Feb 2026 • Self-reported
25.8%
HumanEval+
IFM K2-Horizon-0.9B Hugging Face model card, final checkpoint; reasoning_effort=high, temperature=0.6, top_p=0.95 • Self-reported
79.9%
LiveCodeBench v6
IFM K2-Horizon-0.9B Hugging Face model card, final checkpoint; reasoning_effort=high, temperature=0.6, top_p=0.95 • Self-reported
37.4%
MBPP+
IFM K2-Horizon-0.9B Hugging Face model card, final checkpoint; reasoning_effort=high, temperature=0.6, top_p=0.95 • Self-reported
68.0%

License & Metadata

License
apache_2_0
Announcement Date
September 3, 2026
Last Updated
October 31, 2026

Similar Models

All Models

Recommendations are based on similarity of characteristics: developer organization, multimodality, parameter size, and benchmark performance. Choose a model to compare or go to the full catalog to browse all available AI models.