Laguna S 2.1 is documented as suited to long-running coding and research tasks, including work that spans extended tool use and validation loops. It is also described as useful when an agent needs to hold onto a goal over time, recover from failed approaches, and keep making progress.
It is documented as capable of pursuing meaningful engineering work, including building a working HTML/CSS rendering engine from scratch without human intervention. The model is also presented as suitable for agentic coding in terminal or developer-harness workflows where it plans, edits files, runs code, reads results, and continues for hours.
Poolside also describes it as usable for local deployment on high-memory hardware, including a single NVIDIA DGX Spark. This makes it suitable for teams that want to run high-volume agentic work on hardware they control rather than through metered APIs.
import requests
r = requests.post(
"https://api.aimlapi.com/v1/chat/completions",
headers={"Authorization": "Bearer " + AIMLAPI_KEY},
json={
"model": "poolside/laguna-s-2.1",
"messages": [
{
"role": "user",
"content": "Hello!"
}
]
},
)
print(r.json())
const r = await fetch("https://api.aimlapi.com/v1/chat/completions", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"model": "poolside/laguna-s-2.1",
"messages": [
{
"role": "user",
"content": "Hello!"
}
]
}),
});
console.log(await r.json());
curl -X POST https://api.aimlapi.com/v1/chat/completions \
-H "Authorization: Bearer $AIMLAPI_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"poolside/laguna-s-2.1","messages":[{"role":"user","content":"Hello!"}]}'
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Cached input |
The full 1M context window is included at standard pricing
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
Laguna S 2.1 This page | Agentic coding + long-horizon reasoning | |||
| Reasoning + agents | ||||
| Coding + agents | ||||
| Long-context, multimodal & agentic workflows | ||||
| Multimodal reasoning + agents |
Laguna S 2.1 has a 1,048,576 tokens context window and can return up to 131,072 tokens.
Laguna S 2.1 takes text, image as input and returns text.
Use poolside/laguna-s-2.1 as the model id. Requests go to https://api.aimlapi.com/v1/chat/completions.
Laguna S 2.1 became available on July 21, 2026.
Laguna S 2.1 is priced at input $0.123786 / 1M tokens, output $0.247572 / 1M tokens, cached input $0.0082524 / 1M tokens.
Yes, Laguna S 2.1 can stream responses as they are generated.
Yes, Laguna S 2.1 accepts image input alongside text.
Laguna S 2.1 was built by Poolside.
