Models / laguna-s-2-1

Laguna S 2.1

Poolside's model, with open weights. Forcebench has run it in 1 configuration on the API board. API is models reached through a hosted service; Local, models served on Forcebench’s own hardware.

On the API board

Models reached through a hosted service. Laguna S 2.1 is ranked by its one configuration: the 24th-highest score of 26 models, not clearly different from 2 other models. Scores are within 3 attempts (c@3) with 95% intervals; each configuration is compared with the highest-scoring one by whether their 95% intervals overlap. Every configuration on the board → Cost and speed →

37%
within 3 attempts (c@3), 95% CI 32–42 · thinking on · 24th-highest score of 26 models, not clearly different from 2 other models
3.4k
output tokens per task · median 522, p90 10k
30 s
median per task, every attempt counted, including network and queueing · 90th percentile 335 s
3%
no answer within the 32,768-token budget
Every configuration of Laguna S 2.1 on the API board
ConfigurationRank all configsWithin 3 attempts c@3 · 95% CITokens mean · median · p90No answerLatency median per task, every attempt counted
OpenRouter · FP4 · thinking onhighest score38
37%95% CI 32–42
3.4k522 · 10k3%30 s
Suites, at its highest-scoring configuration
thinking on · OpenRouter · FP4: within 3 attempts (c@3) on each suite with 95% intervals, against the median of the API board's models, each at its highest-scoring configuration.
  • 95% interval
  • API board median
Apex
55%, 95% interval 34–74%
Salesforce APIs
33%, 95% interval 16–56%
CI/CD
41%, 95% interval 22–64%
Salesforce CLI
30%, 95% interval 15–52%
Salesforce docs
6%, 95% interval 1–27%
fflib Enterprise Patterns
17%, 95% interval 6–39%
Flow
17%, 95% interval 6–39%
Governor limits & pushback
53%, 95% interval 32–73%
Lightning Web Components
67%, 95% interval 44–84%
Nonprofit Success Pack
39%, 95% interval 20–61%
Packaging
22%, 95% interval 9–45%
Permissions & access
47%, 95% interval 25–70%
Scratch org definitions
0%, 95% interval 0–18%
SOQL
90%, 95% interval 70–97%
Trigger Actions Framework
39%, 95% interval 20–61%
    Show the numbers
    Per-suite scores of Laguna S 2.1's highest-scoring configuration on the API board, with the board's median
    Suitec@3 95% CIAPI medianTasks
    Apex
    55%95% CI 34–74
    		90%20
    Salesforce APIs
    33%95% CI 16–56
    		72%18
    CI/CD
    41%95% CI 22–64
    		88%17
    Salesforce CLI
    30%95% CI 15–52
    		90%20
    Salesforce docs
    6%95% CI 1–27
    		29%17
    fflib Enterprise Patterns
    17%95% CI 6–39
    		61%18
    Flow
    17%95% CI 6–39
    		69%18
    Governor limits & pushback
    53%95% CI 32–73
    		89%19
    Lightning Web Components
    67%95% CI 44–84
    		100%18
    Nonprofit Success Pack
    39%95% CI 20–61
    		81%18
    Packaging
    22%95% CI 9–45
    		69%18
    Permissions & access
    47%95% CI 25–70
    		87%15
    Scratch org definitions
    0%95% CI 0–18
    		50%18
    SOQL
    90%95% CI 70–97
    		100%20
    Trigger Actions Framework
    39%95% CI 20–61
    		86%18

    In the studies

    Private evaluations

    Want this for your team?

    Azul Labs runs Forcebench privately for two kinds of Salesforce team.

    Choosing a model to run yourself?

    A private benchmark of your shortlist

    Open-weight models at the quantisations and reasoning efforts you could run on infrastructure you control, tested on your kind of Salesforce work, with a written recommendation.

    • Private tasks written from your work only ever run on local models; they are never sent to an API.
    • Re-testing after any change of model, quantisation or engine is available.
    • Typical turnaround is 2–3 weeks; large models take about a day of machine time per configuration.
    Get a private benchmarkHow it works

    Building or buying a Salesforce AI tool?

    A private evaluation of your tool

    Your tool on held-out tasks it has never seen, against API baselines (models reached through a hosted service) and Local ones (served on Forcebench’s own hardware), with a written report on where it holds up and where it fails.

    • It can start as a smaller pilot, about 30 held-out tasks with your tool against two baselines, before a full evaluation.
    • Repeat evaluations after your releases are available.
    • Typical turnaround is 2–3 weeks.
    Ask about an evaluationHow it works

    Fixed fee, quoted up front. The quote depends on the number of tasks and the number of models, tools or configurations tested.

    Or email leo@azl.au · azl.au