AI Benchmark Model
K2 Horizon 0.9B
Benchmark scores, pricing, speed, and model comparisons for K2 Horizon 0.9B.
Key scores
K2 Horizon 0.9B is a foundation model developed by the MBZUAI Institute of Foundation Models, released on September 3, 2026. Designed for specialized tasks, the model holds an intelligence index of 3 and a coding index of 3.4, while its math index remains unknown. Performance metrics for this model include a GPQA score of 0.293, an HLE score of 0.054, a SciCode score of 0.069, and an LCR score of 0.0633333333333333. The model is currently available at no cost, with input, output, and blended pricing all set at $0.00 per 1 million tokens. Specific data regarding output speed and time to first token for K2 Horizon 0.9B are not currently available.
Benchmark results
4 tracked benchmark rows for K2 Horizon 0.9B.
GPQA Reasoning
29.3% SciCode Other
6.9% LCR Other
6.3% HLE Other
5.4%