Models / deepseek-v4-1-flash-1017
DeepSeek V4.1 Flash 1017
DeepSeek's model, with open weights. Forcebench has run it in 1 configuration on the Local board. API is models reached through a hosted service; Local, models served on Forcebench’s own hardware.
On the Local board
Models served on Forcebench’s own hardware. DeepSeek V4.1 Flash 1017 is ranked by its one configuration: the highest score of 16 models, not clearly different from 1 other model. 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 →
| Configuration | Rank all configs | Within 3 attempts c@3 · 95% CI | Tokens mean · median · p90 | No answer | Latency median per task, every attempt counted |
|---|---|---|---|---|---|
| Native FP8, FP4 experts · SGLang · high · 4× NVIDIA GB10, 128 GB each (3× DGX Spark, 1× Gigabyte AI TOP ATOM)highest score | 1 | 76%95% CI 71–80 | 8.9k5.9k · 23k | 3% | 454 s |
Latency depends on the hardware: compare it only between configurations on the same hardware.
- 95% interval
- Local board median
Show the numbers
| Suite | c@3 95% CI | Local median | Tasks |
|---|---|---|---|
| Apex | 90%95% CI 70–97 | 70% | 20 |
| Salesforce APIs | 72%95% CI 49–88 | 42% | 18 |
| CI/CD | 82%95% CI 59–94 | 44% | 17 |
| Salesforce CLI | 95%95% CI 76–99 | 38% | 20 |
| Salesforce docs | 18%95% CI 6–41 | 12% | 17 |
| fflib Enterprise Patterns | 56%95% CI 34–75 | 17% | 18 |
| Flow | 72%95% CI 49–88 | 17% | 18 |
| Governor limits & pushback | 84%95% CI 62–94 | 76% | 19 |
| Lightning Web Components | 100%95% CI 82–100 | 83% | 18 |
| Nonprofit Success Pack | 72%95% CI 49–88 | 56% | 18 |
| Packaging | 67%95% CI 44–84 | 28% | 18 |
| Permissions & access | 87%95% CI 62–96 | 57% | 15 |
| Scratch org definitions | 56%95% CI 34–75 | 11% | 18 |
| SOQL | 95%95% CI 76–99 | 85% | 20 |
| Trigger Actions Framework | 94%95% CI 74–99 | 42% | 18 |
In the studies
- Local vs API: where it lands among the configurations of both boards.
- Pushback: whether it pushed back on requests that break at scale.
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.
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.
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