trl-internal-testing/ci-profile
Updated
AI & ML interests
Internal testing artifact mangement for trl library
Recent Activity
View all activity
qgallouedec
updated a
dataset about 3 hours ago
qgallouedec
published a
dataset about 3 hours ago
sergiopaniego
posted an update about 9 hours ago
Post
73
Repo2RLEnv just shipped TaskSmith + 50 high quality RL envs generated from HF repos 🔨
TaskSmith is a specialized harness that turns a merged PR into a verified RL env
the envs come from HF repos (Transformers, TRL, PEFT, Accelerate, Diffusers), shipped as Harbor tasks you can eval or train on
> code: github.com/huggingface/Repo2RLEnv
> dataset: huggingface.co/datasets/FineEnvs/HF_ML_Tasksmith
TaskSmith is a specialized harness that turns a merged PR into a verified RL env
the envs come from HF repos (Transformers, TRL, PEFT, Accelerate, Diffusers), shipped as Harbor tasks you can eval or train on
> code: github.com/huggingface/Repo2RLEnv
> dataset: huggingface.co/datasets/FineEnvs/HF_ML_Tasksmith
Upload Lfm2VlForConditionalGeneration
1
#2 opened about 9 hours ago
by 
albertvillanova
Upload Lfm2ForCausalLM
#1 opened 1 day ago
by albertvillanova
sergiopaniego
posted an update 1 day ago
Post
2602
ThinkingBox from @microsoft is now available as an OpenEnv env (cc @tuhink 🤗 )!
> ThinkingBox is a sandbox for testing agents on business workflows. it simulates a customer, gives the agent MCP tools over a real database, and at the end checks what changed in that database instead of trusting the agent's last message
> ThinkingBox-Bench is the benchmark built on it: 507 tasks across retail, insurance, travel, banking and consulting
> the OpenEnv env runs each task as an episode in its own isolated backend and returns a pass/fail reward from those checks
https://huggingface.co/blog/microsoft/thinkingbox
> ThinkingBox is a sandbox for testing agents on business workflows. it simulates a customer, gives the agent MCP tools over a real database, and at the end checks what changed in that database instead of trusting the agent's last message
> ThinkingBox-Bench is the benchmark built on it: 507 tasks across retail, insurance, travel, banking and consulting
> the OpenEnv env runs each task as an episode in its own isolated backend and returns a pass/fail reward from those checks
https://huggingface.co/blog/microsoft/thinkingbox
Upload Lfm2ForCausalLM
#1 opened 11 days ago
by albertvillanova
sergiopaniego
posted an update 4 days ago
Upload NemotronHForCausalLM
#1 opened 7 days ago
by albertvillanova
Upload NemotronHForCausalLM
#1 opened 7 days ago
by albertvillanova
Upload NemotronHForCausalLM
#1 opened 7 days ago
by albertvillanova
Upload NemotronHForCausalLM
#1 opened 7 days ago
by albertvillanova
Upload Qwen2_5_VLForConditionalGeneration
#20 opened 9 days ago
by albertvillanova
Upload Qwen2VLForConditionalGeneration
#7 opened 9 days ago
by albertvillanova
Upload GptOssForCausalLM
#4 opened 9 days ago
by albertvillanova
Upload Lfm2VlForConditionalGeneration
#1 opened 9 days ago
by albertvillanova
Upload DeepseekV3ForCausalLM
#4 opened 11 days ago
by albertvillanova
Upload DeepseekV3ForCausalLM
#5 opened 11 days ago
by albertvillanova
Upload Phi3ForCausalLM
#2 opened 14 days ago
by albertvillanova
Upload Phi3ForCausalLM
#2 opened 14 days ago
by albertvillanova