Laguna S 2.1: 118B Free & OpenSource Coding Model Beats Models 13x Its Size
This video breaks down Poolside's Laguna S2.1, a fully open, permissively licensed 118-billion-parameter (8B active) mixture-of-experts coding model that beats models 5-13 times its size on agentic coding benchmarks, and shows exactly how to run it for free via OpenRouter, Open Code, or self-hosting.
AI Stack Engineer11 minTranscript found
Quick learning frame
Read this before watching.
Creative automation uses agents to accelerate production while keeping human taste in story, pacing, selection, and critique.
New playlist item from AI Stack Engineer; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Skill you build: The ability to read and compare frontier coding-model benchmark results (Terminal Bench, SWE-Bench, DeepSWE) to judge real agentic capability, and to pick the cheapest legitimate way to run a given open model between free tier, paid API, and self-hosting.
Watch for the shift from claim to mechanism. The learning value is the point where the transcript reveals a repeatable action, tool boundary, context move, review habit, or artifact.
Concept diagram
Where this video fits.
01Brief
02Source
03Generation
04Selection
05Edit
06Taste Review
Deep lesson
Turn this video into working knowledge.
1,663 cleaned transcript words reviewed across 530 timed caption segments.
Thesis
Laguna S 2.1: 118B Free & OpenSource Coding Model Beats Models 13x Its Size teaches a practical creative automation move: This video breaks down Poolside's Laguna S2.1, a fully open, permissively licensed 118-billion-parameter (8B active) mixture-of-experts coding model that beats models 5-13 times its size on agentic coding benchmarks, and shows exactly how to run it for free via OpenRouter, Open Code, or self-hosting.
The goal is not to remember the video. The goal is to extract the operating principle, tie it to timestamped evidence, test how far the claim transfers, and make something reusable.
0:19
Poolside's origin story
“fully open coding model with a permissive commercial license that competes with models 10 or even 20 times its size, that gets my attention. The model is called Laguna S2.1. It's from a company called Poolside, and right...”
Poolside was founded in 2023 by former GitHub CTO Jason Warner and Eiso Kant, raised about $626 million early on plus a later $1 billion Nvidia commitment pushing its valuation to roughly $12 billion, and bets narrowly that AI for software engineering is the path to more general intelligence; Laguna S2.1 is its third release in under three months, trained in under nine weeks on 4,096 Nvidia H200 GPUs through their internal 'model factory' pipeline. Write a one-paragraph timeline of Poolside's model releases (M1, XS2, XS2.1, S2.1) alongside its funding events in chronological order.
3:07
MoE specs vs benchmarks
“release. Frontier level agentic coding small enough to sit on your desk. On Terminal Bench 2.1 which tests long horizon tasks where the agent works through a terminal S2.1 scores 70.2% For context Tencent's High 3 at 295...”
Laguna S2.1 is a 118-billion-parameter mixture-of-experts model with only about 8 billion active per token (256 routed experts plus one shared, top-10 routing) under the permissive OpenMDW 1.1 license, yet it scores 70.2% on Terminal Bench 2.1 and 78.5% on SWE-Bench Multilingual, beating far larger models like the 1.6-trillion-parameter DeepSeek-V4-Pro-Max. Build a table comparing S2.1, Tencent Hunyuan/High3, and DeepSeek-V4-Pro-Max across Terminal Bench, SWE-Bench Multilingual, and DeepSWE scores to see where the parameter-efficiency gap is largest.
9:51
Free access paths
“their models. So, keep confidential client code away from the free endpoints. The option I'd actually use for coding is open code. Laguna S2.1 is free there through open code zen for a limited time. And unlike the...”
You can run Laguna S2.1 for free through OpenRouter's dedicated endpoint (262K token context) or through Open Code Zen (full 1 million token context, no API key needed), or self-host the Hugging Face weights; Poolside warns that free-tier usage may be used to train future models, so confidential code should stay off it. Install Open Code, run /models, and select Laguna S2.1 free under the zen provider to test it against one of your own repos.
01
Brief
Start with this video's job: This video breaks down Poolside's Laguna S2.1, a fully open, permissively licensed 118-billion-parameter (8B active) mixture-of-experts coding model that beats models 5-13 times its size on agentic coding benchmarks, and shows exactly how to run it for free via OpenRouter, Open Code, or self-hosting. Treat "Brief" as the outcome you are trying to make visible, not a topic label. Anchor it to 0:19, where the video says: “fully open coding model with a permissive commercial license that competes with models 10 or even 20 times its size, that gets my attention. The model is called Laguna S2.1. It's from a company called Poolside, and right...”
02
Source
Use "Source" to locate the part of the creative automation workflow the video is demonstrating. Ask what changes in your real setup if this claim is true. Anchor it to 3:07, where the video says: “release. Frontier level agentic coding small enough to sit on your desk. On Terminal Bench 2.1 which tests long horizon tasks where the agent works through a terminal S2.1 scores 70.2% For context Tencent's High 3 at 295...”
03
Generation
Turn "Generation" into the reusable artifact for this lesson: A creative workflow board with critique criteria and review checkpoints. This is where watching becomes something you can inspect and reuse.
04
Selection
Use "Selection" as the application surface. Decide whether the idea touches a browser flow, a local file, a model choice, a source document, a UI, or a review step.
05
Edit
Use "Edit" to prove the lesson. The evidence should connect back to the video title, transcript anchors, and a concrete output, not a generic best-practice claim.
06
Taste Review
Use "Taste Review" to carry the idea forward: save the prompt, checklist, diagram, or operating rule that would make the next agent run better.
Example
Source-backed work packet
Convert the video into a scoped task that includes the transcript claim, target workflow, acceptance criteria, and proof. The output should be a creative workflow board with critique criteria and review checkpoints..
Example
Claim vs. demo brief
Separate what the speaker claims, what the demo actually proves, and what still needs outside verification before you adopt the workflow.
Example
Teach-back module
Transform the lesson into a definition, a mechanism diagram, one misconception, one practice exercise, and a check-for-understanding question.
Do not learn it wrong
Treating the title as the lesson without checking what the transcript actually says.
Letting the prompt drift into generic advice that could apply to any video in the playlist.
Copying the tool setup without identifying the operating principle that transfers to your own stack.
Skipping the artifact, which means the learning never becomes operational or inspectable.
Do not count this as learned until these are true.
01
State the transcript-backed claim in your own words: This video breaks down Poolside's Laguna S2.1, a fully open, permissively licensed 118-billion-parameter (8B active) mixture-of-experts coding model that beats models 5-13 times its size on agentic coding benchmarks, and shows exactly how to run it for free via OpenRouter, Open Code, or self-hosting.
02
Explain the practical stakes without hype: New playlist item from AI Stack Engineer; queued for transcript-backed review, topic mapping, and a practical learning artifact.
03
Map the idea onto the Brief -> Source -> Generation -> Selection -> Edit -> Taste Review sequence and name the weakest link.
04
Produce the artifact and include the evidence that proves it: A creative workflow board with critique criteria and review checkpoints.
Put it into practice
Give this grounded prompt to Codex or Claude after watching.
You are helping me turn one specific YouTube video into real, durable learning.
Source video:
- Title: Laguna S 2.1: 118B Free & OpenSource Coding Model Beats Models 13x Its Size
- URL: https://www.youtube.com/watch?v=mvqNJbzDexs
- Topic: Creative Automation
- My current learning frame: Install Open Code, select Laguna S2.1 free under the zen provider, give it a real multi-step coding task in your own repo, and compare its persistence and self-verification behavior against your usual coding model.
- Why this matters: New playlist item from AI Stack Engineer; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Transcript anchors from this exact video:
- 0:19 / Evidence 1: "fully open coding model with a permissive commercial license that competes with models 10 or even 20 times its size, that gets my attention. The model is called Laguna S2.1. It's from a company called Poolside, and right..."
- 3:07 / Evidence 2: "release. Frontier level agentic coding small enough to sit on your desk. On Terminal Bench 2.1 which tests long horizon tasks where the agent works through a terminal S2.1 scores 70.2% For context Tencent's High 3 at 295..."
- 6:13 / Evidence 3: "searching the web, finding the original pull request the task was based on, and applying the real fix, which is honestly smart behavior for a coding agent, but it ruins the benchmark. So, they added a prompt telling..."
- 8:15 / Evidence 4: "They also gave the model much more generous rollout budgets during training, longer timeouts, more turns per task, which they think is where the persistence comes from. And they trained rollouts across multiple different agent harnesses, so it..."
- 9:51 / Evidence 5: "their models. So, keep confidential client code away from the free endpoints. The option I'd actually use for coding is open code. Laguna S2.1 is free there through open code zen for a limited time. And unlike the..."
Your task:
1. Use the transcript anchors above as the primary source packet. If you add outside context, label it clearly as outside context and keep it secondary.
2. Create a source-check table with columns: timestamp, claim, what the demo proves, confidence, and what still needs verification.
3. Extract the actual teachable claims from the video. Do not invent claims that are not supported by the title, lesson frame, or transcript anchors.
4. Build a reusable learning artifact: A creative workflow board with critique criteria and review checkpoints.
5. Include:
- a plain-English definition of the core idea
- a diagram or structured model using this sequence: Brief -> Source -> Generation -> Selection -> Edit -> Taste Review
- 3 concrete examples that apply the video idea to real agentic work
- 2 failure modes the video helps prevent
- a checklist I can use the next time I run Codex or Claude
- one practical exercise with a clear done signal
6. Add a "learning transfer" section: what changes in my workflow tomorrow if I actually learned this?
7. Add a "source check" section that cites which transcript anchor supports each major takeaway.
Quality bar:
- Make this specific to "Laguna S 2.1: 118B Free & OpenSource Coding Model Beats Models 13x Its Size", not a generic Creative Automation essay.
- Prefer operational examples, failure modes, and reusable artifacts over broad definitions.
- Call out uncertainty instead of smoothing over weak evidence.
- If evidence is weak, say what transcript segment or timestamp needs review instead of guessing.
- Finish with a concise artifact I could paste into my learning app.
Misconceptions
What to stop believing.
Creative AI removes the need for taste.
It increases the need for taste because output volume explodes.
The best prompt is enough.
References, critique, iteration, and post-production matter just as much.
Practice studio
Learning only counts when you make something.
01
Transcript evidence map
Separate what the video actually says from what you already believe about the topic.
3 source-backed takeaways with timestamps, confidence, and a transfer note.02
One useful artifact
Apply the video to a real workflow and produce a creative workflow board with critique criteria and review checkpoints..
A reusable artifact with a done signal and one verification step.03
Teach-back card
Explain the lesson to someone who has not watched the video yet.
A 90-second explanation, one diagram, one example, and one misconception to avoid.
Recall check
Answer first, then reveal — without rewatching.
Who co-founded Poolside, and what is the company's central bet about the path to general intelligence?
How many of Laguna S2.1's 118 billion total parameters are active per token, and what benchmark shows it beating models 5-13 times its size?
What is one free way to run Laguna S2.1 with its full 1 million token context window?
Source shelf
Use the video as a doorway, then verify with primary sources.
