← Back to Explorer
Pending

Laguna-S-2.1 (2026)

Laguna-S-2.1 is a large language model released by Poolside, released under the openmdw-1.1 license, with a claimed knowledge cutoff of 2026-07-21. Its weights, configuration, and tokenizer files have been cryptographically hashed and are queued for on-chain attestation by Weight Registry.

Description

Laguna S 2.1 is Poolside's coding-agent model, an 118B-total/8B-active MoE released under the OpenMDW-1.1 permissive license and designed for long-horizon agentic coding via the Poolside Agent CLI and OpenAI-compatible APIs. Poolside reports a 40.4% score on DeepSWE, positioning it as outperforming several larger coding-agent models.

Metadata

Type
LLM
Issuer
Poolside
License
openmdw-1.1
Size
218.98 GB
Knowledge cutoff
2026-07-21
Source
https://huggingface.co/poolside/Laguna-S-2.1
Revision
0f573140834b11cfac0c2af97a101a7a69a13e22

Hugging Face stats

71,242 downloads on Hugging Face · 1,031 likes · pipeline: text-generation · library: transformers

transformerssafetensorslagunatext-generationlaguna-s-2.1vllmconversationalcustom_codeeval-results

Weight Hashes

Weights file
46 shards (Merkle root) · 218.98 GB
SHA-256dd66311be6ba10a719378a9e33666e93bfc3c4a285247a7644a4a57cdc39c9aa
SHA3-256001ce20bd0bc244a21295a6aa76410379984ac5ccc30026efe6e8bfa4f092c34
BLAKE322282c2587a84eeeba9a68f68bbefa9ad93e8c2b7693f547bc8b58cbfdc28f08
Config file
config.json · 4.53 KB
SHA-2568309d2ab0da8ac0981b8803b1a4637d843c10fdf7851ddd202ca918fb682392c
SHA3-2564c9b4f0b40e159dea16bc0629d110827e44b208e660c92ba78e05d56824c1022
BLAKE3e0f91701964c2226dc7570a30477533a4c8d698ba9f59b8868f43986230c23d0
Tokenizer file
tokenizer.json · 6.95 MB
SHA-256809240f7a182cde859a4fc4ebc902e619a173d507e99304c1092aa04e7a6658e
SHA3-256a758dbcf604cdc89316935e3e55cb2f26a4ea62048fc17ee5b7dd421a7eea145
BLAKE30424aedd5ed92344e33e8f98f3e01db3a59ae2a76dc72d3b9d51b5a870ac2471

Verify it yourself

Download the file(s) from the source below, then hash them locally (sha256sum <file> on Linux/macOS, certutil -hashfile <file> SHA256 on Windows) and compare the output against the hash(es) shown above.