Preferences and analytics

BenchSift keeps your language, theme, and comparison preferences in your browser. Audience measurements also help improve the site.

Back

Institute of Foundation Models

K2 Horizon 0.9B

3.0 Intelligence Index2026-09-03ReasoningOpen weights
Artificial Analysis Indices
Default
Intelligence3.0
Coding3.4
Agentic0.8
Standard Benchmarks
GPQA Diamond29.3%
HLE5.4%
SciCode6.9%
AA-LCR6.3%
Terminal-Bench 2.11.1%
τ-Bench Banking4.9%
Capabilities
Context window131K tokens
Openness Index53 / 100
Input
text
Output
text
Meta-information
Tokens used (verbosity)60.0M