Models · Text
vikasit-27b
Powerful dense model. Deep reasoning, advanced coding, research tasks.
Parameters
27B dense
Context window
256K
Best for
Research
Benchmarks
MMLU75.2%
GSM8K74.0%
MATH42.3%
HumanEval51.8%
BBH74.9%
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-27b"