git
ask
hub
Privacy
Terms
Sign in with GitHub
saccharomycetes/activevision
↗
Stars ·
12
Language ·
Python
License ·
Apache-2.0
Simplify and Visualize This Repo
Scan the safety of this repo
Self Host this repo
Check out more work of Developer: saccharomycetes
Ask anything about this repo to start.
Using the ActiveVision benchmark repo, write a script that runs the Claude 3.5 Sonnet model on all 17 tasks with a reasoning effort of medium, and saves the graded results to the output directory.
I want to evaluate a Gemini model using the ActiveVision benchmark. How do I pass a model identifier and reasoning effort level to the Gemini evaluation script, and where do I find the per-task accuracy summary?
Help me set up the ActiveVision optional ablation where a coding agent inspects images using code. Show me how to configure and run this tool-use scenario and interpret the resulting accuracy.
I cloned the ActiveVision repo but my evaluation run was interrupted halfway through. How do I resume the run without repeating completed tasks, and how does the grader extract and normalize the model's final answer?
Full explanation on explaingit →
📎
Send
By chatting or signing in you agree to the
Terms
and chat-message logging (revocable in
History
).