Skip to content
Poolside

Pulse. Real-time data from our models usage.

Models usage metrics, refreshed daily from OpenRouter and Hugging Face.

7.4 trillion tokens processed on OpenRouter since our first launch.

*Days since Laguna XS.2 and Laguna M.1 launch. Tokens processed via OpenRouter, all deployments. The total excludes tokens processed through our Direct API and Vercel AI Gateway. Source: OpenRouter, as of 29 Jul 2026 (updated 16:11 UTC).
    • 3 trillion Tokens last 30 days
    • 538 billion Tokens last 7 days
    • 75 billion Tokens last day
    • 171 billion Record daily tokens (30 Jun 2026)

    Stats for all models.

Top models on OpenRouter last month

  • 1st MiMo-V2.5 33T
  • 2nd Hy3 25.1T
  • 3rd DeepSeek V4 Flash 24.5T
  • 4th MiniMax M3 14.9T
  • 5th GLM 5.2 13.7T
  • 6th DeepSeek V4 Pro 12.2T
  • 7th Nemotron 3 Ultra 9.4T
  • 8th Step 3.7 Flash 5.9T
  • 9th Hy3 preview 3.6T
  • 10th MiMo-V2.5-Pro 2.6T
  • 11th gpt-oss-120b 2.2T
  • 12th Laguna M.1 2.1T
  • 13th Kimi K3 2T
  • 14th DeepSeek V3.2 1.7T
  • 15th Nemotron 3 Super 1.6T
  • 16th Gemma 4 31B 1.6T
  • 17th Gemma 4 26B A4B 1.5T
  • 18th Kimi K2.6 1.3T
  • 19th Mistral Nemo 1.1T
  • 20th GLM 5 1.1T
  • 28th Laguna XS 2.1 505B
  • 29th Laguna S 2.1 351B
  • 52nd Laguna XS.2 106B
Tokens over the last 30 days. Source: OpenRouter (openrouter.ai/rankings), as of 29 Jul 2026 (updated 16:11 UTC). View open models leaderboard

Top apps using Poolside models

  • Hermes Agent Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience.
    224B
  • Kilo Code Kilo Code is an open-source AI coding agent that works across VS Code, JetBrains, and CLI to help developers ship code faster with agentic workflows.
    170B
  • Cline Cline is an open-source AI coding agent that lives inside your IDE, autonomously exploring your codebase, editing files, running terminal commands, and using browser automation.
    88B
  • Claude Code Claude Code is Anthropic's agentic coding tool that reads your entire codebase, plans and executes changes across files, runs tests, and iterates on failures, all from natural language prompts.
    39B
  • OpenClaw OpenClaw is an open-source AI agent that connects to your messaging apps and takes real actions on your behalf, from running commands and browsing the web to managing files and sending emails.
    30B
  • Oh-My-Pi
    11B
Tokens across all Laguna deployments, last 30 days. Apps that do not expose token counts on OpenRouter, including OpenCode, are excluded. Source: OpenRouter (openrouter.ai), as of 29 Jul 2026 (updated 16:11 UTC). Source: Laguna XS 2.1 / Laguna S 2.1 / Laguna M.1

Top coding agents on OpenRouter

Tokens over the last 30 days. Source: OpenRouter (openrouter.ai/apps), as of 29 Jul 2026 (updated 16:11 UTC).

1.1M model weight downloads on Hugging Face since our first launch.

Download weights
*Days since Laguna XS.2 and Laguna M.1 launch. Model-repository downloads, grouped by model family. Source: Hugging Face Publisher Analytics, through 28 Jul 2026 (updated 16:11 UTC). Downloads are counted from requests to Hugging Face's library-specific query files (for standard Transformers repositories, config.json); viewing a model card does not count, and automated CI/CD requests may be included. How Hugging Face counts downloads.
    • 705.5k Downloads last 30 days
    • 540.5k Downloads last 7 days
    • 5k Downloads last day
    • 11.5k Avg. downloads per day

    Stats for all models.

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

from openai import OpenAI
import os

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?"
  }]
)

OpenAI-compatible chat API.

Get an API key
$ curl -fsSL https://downloads.poolside.ai/pool/install.sh | sh

poolv1.0.6

• Connected to agent server · Laguna S 2.1

› Audit the checkout flow for race conditions

● Thinking

⎿ Tracing order creation and transaction boundaries

● Read(src/checkout/createOrder.ts)

● Edit(src/checkout/createOrder.ts)

● Bash(pnpm test checkout)

⎿ ✓ 12 tests passed in 284ms

⠹ Working… (12s)

Laguna in your terminal with pool.

CLI docs

$ docker run --gpus all poolside/laguna

NVIDIA DGX Spark · GB10 Grace Blackwell

128 GB unified memory · CUDA device ready

Loading poolside/laguna-s-2.1

Precision FP4 · Context 131,072

Starting OpenAI-compatible inference server...

✓ Ready at http://localhost:8000/v1

Run Laguna locally.

View the local setup guide

Work with Poolside Desktop Assistant.

Download for macOS

Our partners

Gateways

Agents

Local runtimes

Benchmarks

Fine-tuning

Open source

Discover how we built our models.

Start building with Laguna

Bring it into your existing tools and harnesses in minutes.
API keys
xs
2.1
s
2.1
Free in Preview for a limited time.