Models · Text
vikasit-14b
Strong all-rounder. Complex reasoning, long documents, code review.
Parameters
14B
Context window
256K
Best for
Advanced tasks
Benchmarks
GPQA Diamond64.0%
AIME 202479.3%
AIME 202570.4%
LiveCodeBench63.5%
MATH-50096.8%
Thinking mode, no tools where applicable. See full comparisons on the benchmarks page.
Run it
# via Vikasit Inference (OpenAI-compatible)
base_url = "https://api.vikasit.ai"
model = "vikasit-14b"