TMaxxx/tmax-9b-v1.1
Reinforcement Learning • 9B • Updated • 97
Tmax v1.1
Models and datasets for TMax v1.1. Paper: https://arxiv.org/abs/2606.23321. Original release: https://huggingface.co/collections/allenai/tmax
Note RL model based on Qwen3.5-9B.
TMaxxx/tmax-27b-v1.1
Reinforcement Learning • 2.65M • Updated • 103Note RL model based on Qwen3.6-27B.
TMaxxx/tmax-4b-v1.1
Reinforcement Learning • 4B • Updated • 86Note RL model based on Qwen3.5-4B.
TMaxxx/tmax-2b-v1.1
Reinforcement Learning • 2B • Updated • 94Note RL model based on Qwen3.5-2B.
TMaxxx/tmax-8b-v1.1
Reinforcement Learning • 8B • Updated • 82Note RL model based on tmax-sft-8b.
TMaxxx/tmax-13k-v1.1
Viewer • Updated • 13.6k • 2.81kNote Hardened, prefiltered TMax RL training data: 13,565 tasks, formatted for Open Instruct. ODC-BY.
TMaxxx/tmax-13k-v1.1-metadata
Viewer • Updated • 28.2k • 280Note Task details and repair/validation metadata for the same 13,565 training tasks. ODC-BY.
TMaxxx/tmax-13k-v1.1-harbor
Viewer • Updated • 13.6k • 250Note The same 13,565 hardened tasks in Harbor format, with public Docker images. ODC-BY. Run directly from Harbor Hub: https://hub.harborframework.com/datasets/tmax/tmax-13k-v1.1
TMaxxx/qwen35-9b-dppo-endless-terminals-v1.1
Reinforcement Learning • 9B • Updated • 70Note (rerun with new infra)
-
TMaxxx/open-instruct-endless-terminals
Viewer • Updated • 2.49k • 84
TMaxxx/qwen35-9b-dppo-terminal-traj-v1.1
Reinforcement Learning • 9B • Updated • 71Note (rerun with new infra)
-
TMaxxx/open-instruct-terminal-traj
Viewer • Updated • 5.65k • 90
TMaxxx/qwen35-9b-dppo-termigen-v1.1
Reinforcement Learning • 9B • Updated • 65Note (rerun with new infra)
-
TMaxxx/open-instruct-termigen
Viewer • Updated • 3.56k • 96
TMaxxx/qwen35-9b-dppo-swe-smith-v1.1
Reinforcement Learning • 9B • Updated • 75Note (rerun with new infra)
-
TMaxxx/open-instruct-swe-smith
Viewer • Updated • 59.1k • 150 • 1
TMaxxx/qwen35-9b-dppo-cli-gym-v1.1
Reinforcement Learning • 9B • Updated • 65Note (rerun with new infra)
-
TMaxxx/open-instruct-cli-gym
Viewer • Updated • 1.55k • 88
TMaxxx/qwen35-9b-dppo-openthoughts-v1.1
Reinforcement Learning • 9B • Updated • 64Note (rerun with new infra)
-
TMaxxx/agent-task-openthoughts-agent-rl-5k-v1.1
Viewer • Updated • 5k • 1.38k -
TMaxxx/qwen35-9b-dppo-litecoder
Reinforcement Learning • 9B • Updated • 103 -
TMaxxx/agent-task-litecoder-terminal-rl-preview-v1.1
Viewer • Updated • 602 • 3.08k -
TMaxxx/qwen35-9b-dppo-calibforge
Reinforcement Learning • 9B • Updated • 73 -
TMaxxx/agent-task-calibforge-v1.1
Viewer • Updated • 5.43k • 15.3k -
TMaxxx/qwen35-9b-dppo-seta-env
Reinforcement Learning • 9B • Updated • 83 -
TMaxxx/agent-task-seta-env-v1.1
Viewer • Updated • 4.57k • 2.02k -
TMaxxx/qwen35-9b-dppo-terminal-lego
Reinforcement Learning • 9B • Updated • 75 -
TMaxxx/agent-task-terminal-lego-15k-v1.1
Viewer • Updated • 15k • 4.61k -
TMaxxx/qwen35-9b-dppo-facet
Reinforcement Learning • 9B • Updated • 73 -
TMaxxx/agent-task-facet-terminal-6k-v1.1
Viewer • Updated • 6.02k • 3.49k -
TMaxxx/qwen35-9b-dppo-rts
Reinforcement Learning • 9B • Updated • 70 -
TMaxxx/agent-task-recursive-task-synthesis-v1.1
Viewer • Updated • 37.5k • 8.16k