will brown

if there's interest, would love to try and get this working with the [verifiers](https://github.com/willccbb/verifiers) repo i'm building, mostly focused on TRL so far (https://github.com/huggingface/trl/pull/2810) but hopefully the Environments can be...

Sorry, just saw this -- will take a swing when [#53](https://github.com/Blaizzy/mlx-vlm/pull/53) is merged.

the code to implement batch inference is pretty simple, you can just copy the code from mlx_parallm (or probably better: https://github.com/N8python/gsm-mlx) and add it to your codebase. the hard part...

@ParamThakkar123 this doesn't seem like it's doing any actual filtering at the dataset level, just adding fields?

nice! absolutely agree that we want to add VLMs eventually, just hasn't been at top of our priority list yet, though this implementation does look like a pretty nice starting...

Ah we deprecated the report feature, sorry!

@lakshyaag Can this not already be done by passing response_format via sampling_args? Would rather have that be the all-in-one route for expressing additional configurations rather than explicitly adding each one.

ah, I should update the contributor guidance -- developing on Mac without the trainer-only extras is totally fine (I often do this myself), but the trainer-related imports won't be resolved...

nice! looks pretty good, updated to merge with latest main -- probably will make some other edits before merging, our logic for vf-eval outputs json saving has drifted a bit...

nice! would you wanna do the PR on top of the trajectories branch? https://github.com/PrimeIntellect-ai/verifiers/pull/549 there's a bunch of new things which should make native gym-style rollouts much easier, especially for...