Models — Poolside

These models learned to code inside ❯ pool

All Laguna models are trained with reinforcement learning inside our agent harness; they perform at their best in pool, or you can use any ACP client.

curl -fsSL https://downloads.poolside.ai/pool/install.sh | sh

pool — api

S2.1 The daily driver for feature work across a handful of files. Capability you can afford to run all day. ❯ add rate limiting to the public endpoints

Laguna

Weights\↗ API Key

Frontier-class reasoning at mid-size cost. A 118B total parameter MoE model with 8B activated. Trained in-house on 30T tokens.

Our strongest model for agentic coding and long-horizon work. Competitive with models several times its size on Terminal-Bench 2.1 and DeepSWE. Small enough to run on a single NVIDIA DGX Spark.

Terminal-Bench 2.1

Resolved tasks on Terminal-Bench 2.1.

Score
0.0
0.0
0.0
0.0

SWE-Bench Multilingual

Resolved tasks on SWE-Bench Multilingual.

Score
0.0
0.0
0.0
0.0

SWE-Bench Pro (Public Dataset)

Resolved tasks on SWE-Bench Pro (Public Dataset).

Score
0.0
0.0
0.0
0.0
0.0
0.0
0.0

DeepSWE

Resolved tasks on DeepSWE.

Score
0.0
0
0
0.0

SWE Atlas (Codebase QnA)

Resolved tasks on SWE Atlas (Codebase QnA).

Score
0.0
0.0
0.0

Toolathlon Verified

Resolved tasks on Toolathlon Verified.

Score
0.0
0.0
0.0
0.0
0.0

Benchmarks as of 21 July 2026. pass@1 averaged over 4 attempts per task, except DeepSWE, SWE Atlas (Codebase QnA) and Toolathlon Verified that had 3 attempts per task. For all benchmarks we take the maximum of the vendor self-reported score, benchmark author leaderboard or third-party leaderboard (Artificial Analysis), except SWE Atlas (Codebase QnA) where we do not use third-party leaderboard figures.

Benchmark

Model Score
Laguna S 2.1 70.2 at 118B
Laguna XS 2.1 33.4 at 33B
Kimi K3 88.3 at 2800B
DeepSeek-V4-Pro-Max 64 at 1600B
Inkling 63.8 at 975B
Nemotron 3 Ultra 56.4 at 550B
MiniMax M3 66 at 428B
Hy3 71.7 at 295B
DeepSeek-V4-Flash-Max 61.8 at 284B
Inkling-Small 52.7 at 276B
Nemotron 3 Super 38.6 at 120B
Mistral Small 4 21.4 at 119B
Qwen3.6-35B-A3B 44.9 at 35B
Qwen3.6-27B 51.3 at 27B

Total parameters (B, log scale) Score (Pass@1)

Laguna

Weights\↗ API Key

Our lightest and fastest agentic coding model. A 33B total parameter MoE model with 3B activated. Trained in-house on 15T tokens.

SWE-bench Verified

Resolved tasks on SWE-bench Verified.

Score
0.0
0.0
0.0
0.0
0.0
0.0

SWE-bench Multilingual

Resolved tasks on SWE-bench Multilingual.

Score
0.0
0.0
0.0
0.0

SWE-Bench Pro

Resolved tasks on SWE-Bench Pro.

Score
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0

Terminal-Bench 2.0

Resolved tasks on Terminal-Bench 2.0.

Score
0.0
0.0
0.0
0
0.0
0.0
0.0
0.0

Quantize it or serve it.

The weights are yours. Bring it into your existing tools and harnesses in minutes.

We believe the West needs strong open-weight models, and we're committed to contributing to that ecosystem. We release the Laguna family under permissive licences from OpenMDW to Apache 2.0, inviting the community to evaluate, fine-tune, and build on our work.

Get started with Laguna. Free to use for a limited time.

Examples:

Python

import os
from openai import OpenAI

client = OpenAI(
  api_key=os.getenv("POOLSIDE_API_KEY"),
  base_url="https://inference.poolside.ai/v1"
)

response = client.chat.completions.create(
  model="poolside/laguna-s-2.1",
  messages=[
    {
      "role": "user",
      "content": "What are channels in Go?"
    }
  ],
  stream=True
)

for chunk in response:
  if chunk.choices and chunk.choices[0].delta.content is not None:
    print(chunk.choices[0].delta.content, end="", flush=True)

OpenAI-compatible Chat API.

Laguna S 2.1 and XS 2.1 are available on OpenRouter and through our dedicated API, so you can work with your existing tools and harnesses.

Use wherever your work gets done.

For the best experience, use our agent harness, pool, and any ACP-compatible client.

curl -fsSL https://downloads.poolside.ai/pool/install.sh | sh

Get started with NVIDIA DGX Spark.

How we build these models.

The Model Factory, our training infrastructure, and everything we publish as we go.

Footnotes.

All benchmarking for Laguna S 2.1 and Laguna XS 2.1 was completed using the Laude Institute's Harbor Framework with our agent harness, a maximum of 500 steps and sandboxed execution via our internal sandbox service. We report mean pass@1 averaged over multiple attempts per task. Final run trajectories for Laguna S 2.1 evaluation are available to view and download at trajectories.poolside.ai.

Some base task images and verifiers were patched to fix infrastructure reliability issues inherent in task setup, such as rate limits on third-party dependencies in external registries used by the verifier. More details outlining these updates and other findings are available as part of our Laguna M.1/XS.2 Technical Report.

We used the highest publicly-referenced scores for all comparison models across each benchmark.