10 actions in 9 pieces of software · ranked by words, meaning and measured trust · 613 ms
- 1grpo_onesided_5-480demo or tutorialno passport yet
modelrecord8 downloads/month
- text-generation
Task: text-generation
- text-generation
skillorchestra-research/ai-research-skillsrecord
- grpo-rl-training
Expert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model training
- grpo-rl-training
- grpo-rlvr-training
Train reasoning and verifiable-task behavior with GRPO and reinforcement learning from verifiable rewards (RLVR). Use when task success is algorithmically checkable (math, code, tool calls, structured output), when designing GRPO reward functions, or when a GRPO run diverges or reward-hacks.
- grpo-rlvr-training
- 4AV-GRPO Joint Audio-Video Generationdemo or tutorialno passport yet
MCP serverrecord
- av_grpo_joint_av_demo_generate
Generate a synchronized audio-video clip from a text prompt.
- av_grpo_joint_av_demo_generate
modelrecord
- gh200-480gb
Task: GH200 480GB
- gh200-480gb
- 6holoscriptfound in code, not yet checkedno passport yet
MCP serverbrianonbased-dev/holoscript
- holo_run_grpo_pass
Run a single GRPO training pass: extract prompts from source files,
- holo_run_grpo_pass
- 7utauMCPfound in code, not yet checkedno passport yet
- add_voice_part
既存の.ustxプロジェクトに新しいボイスパート(ノートを入れる箱)を追加して保存する。 position/durationはtick単位(480=1拍)。
- convert_beats_ticks
拍(四分音符=1.0)とtick(480/拍)を相互変換する。どちらか一方を指定する。
- add_voice_part
- 8TraceHunt MCP Serverfound in code, not yet checkedno passport yet
- hunt_username
Search for a username across 480+ platforms (OSINT).
- hunt_username
- 9cruciblefound in code, not yet checkedno passport yet
MCP servereren23/crucible
- tree_expand_grpo
GRPO-style expansion: keep top-K candidates by group-relative advantage.\n\n
- tree_expand_grpo
Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.