AI Benchmark Model
K2 Horizon 7B
Benchmark scores, pricing, speed, and model comparisons for K2 Horizon 7B.
Key scores
K2 Horizon 7B is an artificial intelligence model developed by the MBZUAI Institute of Foundation Models, released on September 3, 2026. The model is currently offered with no associated costs, featuring input, output, and blended pricing of $0.00 per 1 million tokens. Performance metrics for the model include an intelligence index of 21 and a coding index of 38.6, while the math index remains unknown. Benchmark results for K2 Horizon 7B are recorded as 0.753 on GPQA, 0.182 on HLE, 0.275 on SciCode, and 0.696666666666667 on LCR. Specific data regarding output speed and time to first token are currently unavailable for this model.
Benchmark results
4 tracked benchmark rows for K2 Horizon 7B.
GPQA Reasoning
75.3% LCR Other
69.7% SciCode Other
27.5% HLE Other
18.2%