Nous Research 🌐
DoomBench currently associates 0 models and 3 evidence assessments with Nous Research. Company attribution is separate from each model's version-specific risk score.
- Tracked models
- 0
- Evidence items
- 3
- Toward-doom share
- 0.9%
- Net index pressure
- +0.61
0 comments · 0 votes
Sign in to join the discussion →
No comments yet. Start the discussion.
Models from Nous Research
Availability distinguishes public artifacts, proprietary hosted access, internal systems, and profiles whose current source evidence remains inconclusive.
No version-specific model profiles are associated with this company yet.
Latest assessments involving Nous Research
Hermes adds durable multi-agent work with retries and recovery
Hermes Agent v0.13.0 introduced a persistent Kanban board where multiple agents claim, hand off, and complete tasks using heartbeats, reclaim, zombie detection, retry budgets, incomplete-exit blocking, and a hallucination gate. A goal mode kept an agent focused across turns, and gateway sessions resumed after restarts.
Hermes ships an autonomous curator for its own skills and memory loop
Hermes Agent v0.12.0 added a background Curator that periodically grades, consolidates, and prunes the agent's skill library and writes audit reports. Its separate review process also decides after each turn what memories or skills to save or update, while being restricted to memory and skill tools with protections for pinned and bundled skills.
Nous Research releases Hermes as a multi-platform agent environment
Hermes Agent v0.2.0 packaged a multi-platform messaging gateway, native MCP support, more than 70 skills, IDE integration, isolated parallel Git worktrees, configurable subagents, cross-session memory, browser and sandbox tools, background processes, cloud execution, and rollback checkpoints.