← The Wire

Top 5 · 2026-04-29 · source-backed

Poolside Ships a 68.2% SWE-bench Coding Agent That Runs on Your Mac. Apache 2.0.

Confidence
source-backed
Sources
1
Redaction
passed

Story

Poolside AI released two models that change the math on local coding agents. Laguna M.1 is a 225B total / 23B active MoE model scoring 72.5% on SWE-bench Verified. Laguna XS.2 is a 33B total / 3B active model scoring 68.2% on the same benchmark, 44.5% on SWE-bench Pro, and 30.1% on Terminal-Bench 2.0. Both are Apache 2.0. XS.2 runs locally on a Mac with 36GB RAM via Ollama. (Source)

Alongside the models, Poolside shipped "pool," a lightweight terminal-based coding agent that runs entirely on your machine. It's the same environment they use internally for agent RL training, now available as a research preview. Combine pool with XS.2 and you've got a competitive agentic coding setup with zero API costs and zero data leaving your machine.

I want to put this in context. A year ago, getting above 50% on SWE-bench Verified with a local model felt aspirational. Now we've got 68.2% from a model with 3B active parameters. That's significant. Not because it matches Claude or GPT on the hardest problems, but because it crosses a threshold where local inference becomes genuinely useful for a huge chunk of daily coding work.

The timing matters too. With GitHub Copilot moving to usage-based billing (see story #1), the cost of cloud-hosted coding assistance is about to spike. Having a local alternative that clears the 65% SWE-bench line gives developers a real escape valve. Private codebases, air-gapped environments, developers in regions with spotty API access, indie builders watching their API bills. This serves all of them.

The broader pattern is worth noting. SWE-bench Verified has become the universal legibility benchmark for coding models. Every new launch leads with it. Poolside, Devstral, Qwen Coder, NVIDIA Nemotron. SWE-bench is to coding models what ImageNet was to vision. The standardization makes comparison tractable, but I worry it also incentivizes teaching to the test.

If you've got a Mac with 36GB+ RAM, download XS.2 through Ollama and try pool on a real codebase this week. It's free, it's Apache 2.0, and it might change how you think about paying for AI coding tools.


Related stories

Each link below shares sources, entities, or timing with this story.

  1. Shared entities / Shared topic / Earlier coverage / Tension

    Kimi K2.6: A 1-Trillion-Parameter Open-Weights Model Just Beat GPT-5.4 on SWE-Bench Pro

    Both cover Bench, GPT, MoE, SWE; overlapping topics (agent, coding, model, swe-bench); earlier Bench coverage from 2026-04-21.

  2. Shared entities / Shared topic / What happened next

    Alibaba open-sources Qwen-AgentWorld, a flight simulator for your agents

    Both cover Apache, GPT, MoE, SWE; overlapping topics (agent, apache, environment, model); picks up the Apache thread on 2026-06-27.

  3. Shared entities / Shared topic / What happened next / Tension

    Cohere open-sourced a 30B coding model that runs on a single H100. Agentic coding without the frontier bill.

    Both cover Apache, Bench, SWE, Terminal; overlapping topics (agent, apache, coding, local, model); picks up the Apache thread on 2026-06-10.

  4. Shared entities / Shared topic / Earlier coverage

    Berkeley Researchers Score 100% on SWE-bench Verified Without Solving a Single Task. Every Major AI Benchmark Is Broken.

    Both cover Bench, Claude, SWE, Terminal; overlapping topics (agent, coding, model, swe-bench); earlier Bench coverage from 2026-04-12.

  5. Shared entities / Shared topic / What happened next

    Devstral 2 is open source, 123B params, hits 72.2% SWE-bench, and costs a seventh of Sonnet

    Both cover Apache, Devstral, SWE, Verified; overlapping topics (agent, apache, coding, cost, model); picks up the Apache thread on 2026-06-02.

  6. Kimi K2.7-Code: a 1T-param open coding model with benchmarks nobody can check

    Both cover Bench, MoE, SWE, Terminal; overlapping topics (coding, model, swe-bench); picks up the Bench thread on 2026-06-14.

  7. Shared entities / Shared topic / Earlier coverage

    DeepSeek V4 Drops: MIT-Licensed Frontier Model at 1/6th the Cost, Running on Chinese Chips

    Both cover Bench, Claude, MoE, SWE; overlapping topics (active, cost, model); earlier Bench coverage from 2026-04-26.

  8. Shared entities / Shared topic / Earlier coverage / Tension

    A 27B Model Just Matched the 400B One. It Fits on Your Laptop.

    Both cover MoE, Ollama, SWE, Verified; overlapping topics (coding, cost, model, swe-bench); earlier MoE coverage from 2026-04-23.

Source trail

Entities

Provenance

AI generated
no
Story unit
2026-04-29-poolside-ships-a-68-2-swe-bench-coding-agent-that-runs-on-your-mac-apache-2-0
Labels
source-backed, canonical briefing excerpt