Models / nemotron-3-ultra

Nemotron 3 Ultra

NVIDIA'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. Nemotron 3 Ultra is ranked by its one configuration: the 20th-highest score of 26 models, not clearly different from 5 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 →

59%
within 3 attempts (c@3), 95% CI 54–64 · thinking on · 20th-highest score of 26 models, not clearly different from 5 other models
1.8k
output tokens per task · median 1k, p90 3.7k
55 s
median per task, every attempt counted, including network and queueing · 90th percentile 317 s
0%
no answer within the 32,768-token budget
Every configuration of Nemotron 3 Ultra on the API board
ConfigurationRank all configsWithin 3 attempts c@3 · 95% CITokens mean · median · p90No answerLatency median per task, every attempt counted
OpenRouter · thinking onhighest score33
59%95% CI 54–64
1.8k1k · 3.7k0%55 s
Suites, at its highest-scoring configuration
thinking on · OpenRouter: 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
75%, 95% interval 53–89%
Salesforce APIs
61%, 95% interval 39–80%
CI/CD
71%, 95% interval 47–87%
Salesforce CLI
60%, 95% interval 39–78%
Salesforce docs
24%, 95% interval 10–47%
fflib Enterprise Patterns
33%, 95% interval 16–56%
Flow
22%, 95% interval 9–45%
Governor limits & pushback
79%, 95% interval 57–91%
Lightning Web Components
94%, 95% interval 74–99%
Nonprofit Success Pack
56%, 95% interval 34–75%
Packaging
39%, 95% interval 20–61%
Permissions & access
67%, 95% interval 42–85%
Scratch org definitions
28%, 95% interval 13–51%
SOQL
95%, 95% interval 76–99%
Trigger Actions Framework
83%, 95% interval 61–94%
    Show the numbers
    Per-suite scores of Nemotron 3 Ultra's highest-scoring configuration on the API board, with the board's median
    Suitec@3 95% CIAPI medianTasks
    Apex
    75%95% CI 53–89
    		90%20
    Salesforce APIs
    61%95% CI 39–80
    		72%18
    CI/CD
    71%95% CI 47–87
    		88%17
    Salesforce CLI
    60%95% CI 39–78
    		90%20
    Salesforce docs
    24%95% CI 10–47
    		29%17
    fflib Enterprise Patterns
    33%95% CI 16–56
    		61%18
    Flow
    22%95% CI 9–45
    		69%18
    Governor limits & pushback
    79%95% CI 57–91
    		89%19
    Lightning Web Components
    94%95% CI 74–99
    		100%18
    Nonprofit Success Pack
    56%95% CI 34–75
    		81%18
    Packaging
    39%95% CI 20–61
    		69%18
    Permissions & access
    67%95% CI 42–85
    		87%15
    Scratch org definitions
    28%95% CI 13–51
    		50%18
    SOQL
    95%95% CI 76–99
    		100%20
    Trigger Actions Framework
    83%95% CI 61–94
    		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