OpenThoughts

open-thoughts's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Articles

penfever 

in open-thoughts/TaskTrove 10 days ago

CrossCodeEval: restore context, benchmark verifier, drop uninferable tasks

#3 opened 10 days ago by

FWeindel

ryanmarten 

authored 3 papers 11 days ago

Harbor Adapters and Harbor-Index: Infrastructure and a Curated Meta-Dataset for Large-Scale Agentic Evaluation

Paper • 2609.04298 • Published 20 days ago

OpenThoughts-Agent: Data Recipes for Agentic Models

Paper • 2606.24855 • Published Jun 23 • 49

Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces

Paper • 2601.11868 • Published Jan 17 • 38

penfever 

updated a dataset 17 days ago

open-thoughts/TaskTrove

Viewer • Updated 10 days ago • 1.91M • 10.1k • 29

penfever 

in open-thoughts/TaskTrove 22 days ago

Vacuous verifiers in TaskTrove — three mechanisms causing reward 1.0 on an untouched workspace (13% of sampled tasks in 11 subsets)

1

#2 opened 2 months ago by

anismahmahi

penfever 

authored a paper 3 months ago

OpenThoughts-Agent: Data Recipes for Agentic Models

Paper • 2606.24855 • Published Jun 23 • 49

RZ412 

authored a paper 3 months ago

OpenThoughts-Agent: Data Recipes for Agentic Models

Paper • 2606.24855 • Published Jun 23 • 49

EtashGuha 

updated a collection 4 months ago

OpenThinker-Agent2

Collection

OpenThinker-Agent2: agentic SFT/RL datasets and 8B/32B models (cold-start SFT, RL, and the OpenThinkerAgent-32B release). • 11 items • Updated Jun 11 • 11

penfever 

updated a collection 4 months ago

OpenThinker-Agent-Complete

Collection

OpenThinkerAgent-32B SFT data-scaling ladder (models + matching datasets, 316->100K) plus TaskTrove & AgentTrove sources. • 15 items • Updated Jun 10 • 5