Maxime Labonne Shares Talk on Designing and Post-Training Edge Agentic Models
maximelabonne · x · 2026-07-31
Maxime Labonne recorded a 30-minute presentation for the Cohere Labs Summer School focused on designing and post-training edge agentic models.
- Core Techniques: Introduces Multi-Domain On-Policy Distillation (MOPD) and agentic reinforcement learning.
- Materials: Shared a comprehensive 40-slide deck alongside the talk.
- Engagement: The Twitch broadcast attracted close to 1,000 viewers and numerous technical questions.
Related event: Cohere Summer School Talk on Designing Edge Agent Models(2 posts)→
More from Models
- KOL on Model Competition: Compute and Resources Rule the Game — teortaxesTex · 2026-07-31
- Tencent's Hy-MT2 Hits 700K Downloads, Releases 30B GGUF for Local Inference — victormustar · 2026-07-31
- Gemini Flash's Low Pricing Hailed as Another 'DeepSeek Moment' — eyishazyer · 2026-07-31
- DeepSeek demonstrates autonomous subagent orchestration without prompts — teortaxesTex · 2026-07-31
- DeepSeek V4's fully compressed layers might bottleneck long-context performance — bookwormengr · 2026-07-31
- Claude 3.5 Sonnet Glitches When Context Limit Exceeded — Numerous-Recover7685 · 2026-07-31