Poolside
150 posts
@poolsideai
We build models for agentic coding and long-horizon tasks. Try Laguna: https://t.co/yVvIYPjMf5
@poolsideai recently released Laguna XS 2.1, an awesome 33B coding model with a DFlash speculative-decoding drafter. Lucebox now runs the pair on a single RTX 3090: - 296 tok/s peak at short context - A flat 152 tok/s at 256K tokens, where the full KV cache would not even fit in 24 GB - ~3,500 tok/s prefill, processing 256K tokens in just 67 seconds Three optimizations got the same GPU from 22 to 152 tok/s at 256K in one pass: a drafter KV ring cache, sliding-window ring caches, and KVFlash paging. And the speculative decoding is lossless: every committed token is exactly one the model itself would have produced. Super proud to support @poolsideai on their work to become the leading western open-source lab. Hope you enjoy it!
Researchers from Berkeley and Princeton are partnering with Eigen Labs to launch a suite of open science autoresearch challenges together on Frontier CS. The paper is being presented at @icmlconf in Seoul today. If you’re there, join the researchers at Hall A 502 from 2:30-4:15 PM local time to discuss. The challenge is live globally: openfrontiercs.com
Big day for Ollama! When we started, open models and the open source AI ecosystem were in their early days with few believers. Our belief in open source has never wavered. With today's fundraising announcement and our 9M+ active builders, we’re ready to scale open models into AI that you can own. All aboard open models! 🧵
Laguna XS 2.1 performed on Qwen 3.6 35B's level in Tetris building and ran 2x faster We tested two open models on a single RTX 3090 in the @poolsideai coding agent. The task was building a playable retro Tetris as one self-contained html file. Each model wrote and rewrote the game across 3 iterations Outputs: Laguna XS 2.1: 45K tokens, 158 tok/s Qwen 3.6 35B: 39K tokens, 81 tok/s The two Tetris builds are near identical. Poolside's Laguna has a couple of small visual bugs that Qwen 3.6 35B doesn't, but it built the same game twice as fast by its built-in DFlash speculative decoding
Today we’re releasing Laguna XS 2.1. It’s a small upgrade to the Laguna XS.2 model, the same 33B total / 3B active MoE and stronger results on multilingual coding and terminal-style tasks. Available now on @huggingface, @OpenRouter, and via Poolside API.
Today we’re releasing Laguna XS 2.1. It’s a small upgrade to the Laguna XS.2 model, the same 33B total / 3B active MoE and stronger results on multilingual coding and terminal-style tasks. Available now on @huggingface, @OpenRouter, and via Poolside API.
Today we’re releasing Laguna XS 2.1. It’s a small upgrade to the Laguna XS.2 model, the same 33B total / 3B active MoE and stronger results on multilingual coding and terminal-style tasks. Available now on @huggingface, @OpenRouter, and via Poolside API.
We're adopting the Linux Foundation’s OpenMDW framework across our open model families. This helps make open model licensing simpler and more consistent at scale. A single legal framework across models, code, documentation, and data helps reduce friction for developers and enterprises building with open source.
