Laguna S 2.1

Fiche modèle · Poolside · laguna-s-2.1

Laguna S 2.1 is Poolside's most capable Laguna 2.1 model: an open-weight Mixture-of-Experts coding model with 118B total parameters and 8B activated per token (256 routed experts + 1 shared; 48 layers). It targets long-horizon and agentic coding with interleaved thinking, a 1,048,576-token context window in thinking and no-thinking modes, and text-to-text only modalities. Available as open weights on Hugging Face and via Poolside's API, OpenRouter, and other hosts (poolside/laguna-s-2.1).

Identité

OrganisationPoolside
Pays—
Famille—
Typellm
Date de sortie2026-07-21
Licenceopenmdw license v1.1
Licence commercialeoui
Licence open sourceoui
Poids ouvertsoui
Auto-hébergement (DGX Spark)non — 118 Md > 107 Md
Modalitéstext
Fenêtre de contexte1 048 576
KV cache / token (Q8)98.3 ko/token
Paramètres totaux118 B
Paramètres actifs (session)8 B
Source des paramètres actifsweb : https://huggingface.co/poolside/Laguna-S-2.1
Curation des paramètres actifs2026-11-04
Prix par session0.14 $ (1 000 000 tokens entrée + 200 000 sortie, prix le moins cher par fournisseur)
Tokens / tâche (Data Players)1.2 M
Tokens / tâche (Terminal-Bench)83.3 M (est.)
Ratio Terminal-Bench / Data Players69.4×
Coût / tâche (Terminal-Bench)1.18 $ (est.)
Tokens d'entraînement—
Coupure de connaissance—
Créé le2026-08-14T16:26:20.435207Z
Mis à jour le2026-11-04T06:05:04.651671Z
SourceLLM Stats
URLhttps://llm-stats.com/models/laguna-s-2.1

Prix par fournisseur

Statut du modèle (offres + auto-hébergement) : non — hors DGX Spark (> 107 Md).

Toutes sources confondues : catalogue llm-stats, routeurs EUrouter/Cortecs et catalogues directs (OVHcloud, Scaleway, Infomaniak). Les prix sont convertis en USD pour la comparaison ; le prix publié et le taux BCE utilisé sont dans l'infobulle. Les prix inconnus restent indiqués « — ».

FournisseurSourceSouverainetéEntrée (/M tok)Sortie (/M tok)QuantificationContexteLatence (s)Débit (tok/s)Statut
Poolsidellm-statsnon0.10 $0.20 $————active

Souveraineté qualifiée par offre (modèle × fournisseur) ; survolez le badge pour les critères.

Souveraineté

Scores par catégorie (top_scores)

CatégorieScore
reasoning0.60
general0.60
agents0.50
code0.60
tool_calling0.60

Scores par benchmark

BenchmarkCatégorieScoreMaxNormalisé (API)VérifiéAuto-déclaréDaté du
deepswe-1.1code0.401.000.40nonoui2026-11-04T06:05:04.677985Z
swe-atlas-codebase-qnacode0.461.000.46nonoui2026-11-04T06:05:04.681073Z
swe-bench-multilingualreasoning0.791.000.79nonoui2026-11-04T06:05:04.662642Z
swe-bench-proreasoning0.591.000.59nonoui2026-11-04T06:05:04.668508Z
terminal-bench-2.1reasoning0.701.000.70nonoui2026-11-04T06:05:04.659527Z
toolathlonagents0.501.000.50nonoui2026-11-04T06:05:04.694327Z

Classements

CatégorieRangScoreNote conservatriceBenchmarks évaluésMéthodeDaté duProvenance
agents69—23.045trueskill2026-10-19model_history
code63—29.445trueskill2026-10-19model_history

Lecture : Note conservatrice = borne basse TrueSkill (niveau estimé − marge d’incertitude) ; Benchmarks évalués = nombre de benchmarks de la catégorie pris en compte. À performances égales, un modèle évalué sur moins de benchmarks obtient une note plus basse : la note ne se déduit pas d’un seul score. Provenance : rankings = top 50 de /rankings ; model_history = historique du modèle (/models/{id}/history), utilisé pour les rangs au-delà du top 50.

Données

Dates de curation (récupération depuis l'API) : models 2026-11-04 · benchmarks 2026-11-04 · scores 2026-11-04 · updates 2026-11-04 · account 2026-11-04 · model_history 2026-11-04 · rankings 2026-11-04