Laguna S 2.1 packs 118B parameters, 8B active, and a 1M-token context
UsedMorning9886 · reddit · 2026-07-22
Poolside released Laguna S 2.1, a 118B-parameter MoE model with about 8B active parameters per token.
The post highlights several claims: it is designed for agentic coding and long-horizon work, uses 256 routed experts plus one shared expert, supports a 1M context window, and includes native reasoning with interleaved thinking between tool calls. Benchmarks cited in the post show it outperforming much larger models such as Nemotron 3 Ultra, DeepSeek V4 Pro Max, and Inkling on several coding and terminal tasks, while still trailing some top-end frontier models. The author also notes an OpenMDW 1.1 license and practical memory requirements: BF16 weights are around 236GB, but Q4 GGUF quants are available.
Related event: Poolside Releases 118B Open-Weight Coding Model Laguna S 2.1(35 posts)→
More from coding & agent
- Agent Field Kit launches $29 security runbooks for production AI agents — alvelda · 2026-10-27
- Dev mixes Claude specs with abliterated GLM to build SS7 threat detector — l4rz · 2026-10-27
- Cloudflare Durable Objects validated as top architecture for AI agent WebSocket servers — tristanbob · 2026-10-27
- GitHub Next's Chopin prototype explores real-time multiplayer agentic planning — max__drake · 2026-10-27
- Veynod: An Open-Source Shared Workspace Where AI Agents Collaborate Like a Team — TKOWNS · 2026-10-27
- TMLR paper: enforced thinking doesn't reliably make LLM agents disclose key info to users — SharonYixuanLi · 2026-10-27