Models · Text
vikasit-4b
Balanced small model. Good code completion and multi-turn chat.
Parameters
4B
Context window
128K
Best for
Code & conversation
Benchmarks
MMLU-Pro74.0%
GPQA Diamond65.8%
AIME 202581.3%
LiveCodeBench55.2%
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-4b"