Poolside Laguna M.1 launches free on OpenRouter with 128k context
TL;DR
Poolside's Laguna M.1 coding model is now free to use on OpenRouter, with a 128k context window and zero per-token cost. The Poolside coding-assistant product itself remains enterprise-only (on-prem or VPC deployment).
What changed
Poolside's Laguna M.1 model is listed at $0 input and $0 output on OpenRouter, with a 131k context window and tool calling. A smaller sibling, Laguna XS.2, is also free on the same platform.
Important scope: this is the model weights served via OpenRouter. The Poolside coding-assistant product (the IDE-integrated agent the company sells to enterprises) is not free; it is deployed on-prem or in a customer VPC and priced for fortune-tier customers.
Why it matters
Developers can now test Poolside's frontier coding model from any OpenRouter-compatible client without signing an enterprise contract. That is a meaningful change in access, even though it does not unlock the full Poolside product surface (workstations, eval harness, secure deployment).
For most coders, the practical question is whether Laguna M.1 holds up against Claude, GPT, and DeepSeek on real tasks. The free tier on OpenRouter is the cheapest way to find out.
What to watch for
Track real-world coding-benchmark performance on SWE-bench and Aider. Watch whether OpenRouter imposes rate limits as usage scales. Note any move by Poolside to open up direct API access beyond OpenRouter.
Who this matters for
- Vibe Builders: Integrate this free coding agent to automate boilerplate generation and speed up prototype iteration.
- Developers: Deploy Laguna M.1 for complex refactoring tasks and agentic workflows without incurring token costs.
Harsh’s take
Poolside is using OpenRouter as a free distribution channel for its model weights while keeping the actual product (the IDE-integrated coding assistant deployed in customer VPCs) firmly enterprise-only. That is a smart play: developer mindshare without diluting the enterprise sales motion.
For Vibe Builders and individual devs, the headline is narrower than it looks. You get the model on OpenRouter; you do not get the Poolside product. Treat Laguna M.1 as a model to A/B test against Claude Sonnet and GPT in your existing tooling, not as a Poolside trial.
If Laguna M.1 actually outperforms Claude or GPT on real coding tasks, this becomes interesting. If it does not, it is just another free model on a long list. The primary value is optionality: zero cost to find out either way.
by Harsh Desai
About OpenRouter
View the full OpenRouter page →All OpenRouter updatesGo deeper
More AI news
- Daily RoundupHugging Face trends with Ideogram LoRA and LFM embeddings, Replicate adds image model, and Pixlie video tools
Hugging Face hosts three new trending models for images and embeddings while Replicate and Product Hunt add practical inference and video options, and industry voices remind users that chatbots lack real relationships.
- FeatureNobel laureate John Jumper is leaving DeepMind for rival Anthropic
John Jumper is leaving DeepMind for Anthropic. Other high-profile researchers are also exiting the lab.
- FeatureHermes-agent supports editing existing images via the image_generate tool
Hermes-agent adds image-to-image edits to the image_generate tool. Users pass a source image and prompt to modify images across supported backends.