Back
FR

Institute of Foundation Models

K2 Horizon 3.7B

ReasoningOpen weights
Intelligence
15.6
Rank 313 of 768 measured models
Coding
26.1
Rank 196 of 299 measured models
Context window
524K
Rank 589 of 1296 measured models

K2 Horizon 3.7B scores 15.6 on the Artificial Analysis Intelligence Index, ranking 313th of 768 measured models. The closest models on this index are Qwen3 Max (15.6), Gemini 2.5 Flash (15.5), and Claude 3.7 Sonnet (15.3).

Artificial Analysis Indices
Intelligence15.6
Rank 313 of 768 measured models
Coding26.1
Rank 196 of 299 measured models
Agentic16.5
Rank 105 of 190 measured models
Capability indexes
Performance in specific industries, measured by Artificial Analysis
Finance & Accounting16.5
Rank 188 of 241 measured models
Strategy & Ops14.6
Rank 182 of 241 measured models
Legal17.6
Rank 186 of 241 measured models
Engineering20.2
Rank 180 of 240 measured models
Economics25.9
Rank 178 of 248 measured models
Standard Benchmarks
GPQA Diamond69.2%
Rank 332 of 620 measured models
HLE13.9%
Rank 285 of 681 measured models
SciCode22.0%
Rank 240 of 248 measured models
AA-LCR62.3%
Rank 291 of 592 measured models
Terminal-Bench 2.128.1%
Rank 167 of 245 measured models
Terminal-Bench 4.00.0%
Rank 170 of 240 measured models
τ-Bench Banking19.6%
Rank 104 of 211 measured models
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy