Salesforce APIs

Integration work against Salesforce's APIs: REST sObject, query, composite, composite graph and sObject Collections resources, Bulk API 2.0 ingest and query jobs, Tooling API, User Interface API, platform events, custom Apex REST endpoints and OAuth 2.0 token requests. Answers are raw HTTP requests, checked deterministically for method, path, query, headers and body (JSON, form-encoded or CSV), plus a few questions on API behaviour and choice.

18
tasks
Deterministic validators
grading
78%
best pass@1 · DeepSeek V4.1 Flash
16
configs with results (4 partial)

How it's graded

Answers are raw HTTP requests, checked deterministically for method, path, query, headers and body (JSON, form-encoded or CSV). A few questions cover API behaviour and choosing the right API.

Difficulty mix

6 easy7 medium5 hard

Who's best at Salesforce APIs

pass@1 on this suite only, with 95% confidence intervals. With 18 tasks the intervals are wide, so read overlapping bars as ties. Open in the full leaderboard →

Salesforce APIs: pass@1 by config with 95% confidence intervals
#Model and configScore pass@1 · 95% CITasks
1DeepSeek V4.1 Flash EXL3 2.9bpw · vLLM + ExLlamaV3 · effort high
78%95% CI 56–94
18
2Gemma 4 31B QAT W4A16 · vLLM · effort on
50%95% CI 28–72
18
3Qwen3.8 Flash-Next NVFP4 · vLLM · effort medium
44%95% CI 22–67
18
4Qwen3.8 27B AWQ-INT4 · vLLM · effort medium
41%95% CI 20–61
18
5Gemma 4 26B-A4B NVFP4 · vLLM · effort on
39%95% CI 17–61
18
5Qwen3.8 27B MLX 8-bit · MTPLX · effort low
39%95% CI 17–61
18
5Qwen3.8 27B MLX 4-bit · MTPLX · effort low
39%95% CI 17–61
18
5Qwen3.8 27B Splash 4-bit · Splash · effort low
39%95% CI 17–61
18
5Qwen3.8 Flash-Next MLX 4-bit · MTPLX · effort medium
39%95% CI 17–61
18
10Qwen3.6 35B-A3B NVFP4 · vLLM · effort on
33%95% CI 11–56
18
10Qwen3.8 27B AWQ-INT4 · vLLM · effort xhigh
33%95% CI 11–56
18
12Qwen3.8 27B AWQ-INT4 · vLLM · effort low
28%95% CI 11–50
18
–GLM-5.3 Flash partialEXL3 4.0bpw · vLLM + ExLlamaV3 · effort maxRun so far: 2 of 15 suites · 27 of 272 tasks graded
73%95% CI 47–93
15
–GLM-5.3 Flash partialEXL3 4.0bpw · vLLM + ExLlamaV3 · effort offRun so far: 2 of 15 suites · 35 of 272 tasks graded
56%95% CI 33–78
18
–GLM-5.3 Flash partialEXL3 4.0bpw · vLLM + ExLlamaV3 · effort highRun so far: 14 of 15 suites · 219 of 272 tasks graded
39%95% CI 17–61
18
–GLM-5.3 Flash partialEXL3 4.0bpw · vLLM + ExLlamaV3 · effort lowRun so far: 2 of 15 suites · 35 of 272 tasks graded
39%95% CI 17–61
18

What's in the suite

Solve rate is the mean pass@1 on that task across every finished run, a rough guide to how hard models find it.

Salesforce APIs tasks with difficulty and solve rate
TaskDifficultySolve rate
Close an opportunity and delete a duplicate contactapi-sobject-update-deleteeasy
100%
Create a Case with the REST APIapi-sobject-create-caseeasy
100%
Pick the API for a nightly 5-million-row loadapi-choose-bulk-for-large-loadeasy
100%
Pre-sync checks with describe and limitsapi-describe-and-limitseasy
92%
Publish a platform event from an external systemapi-platform-event-publisheasy
83%
Query Apex code coverage with the Tooling APIapi-tooling-coverage-queryeasy
8%
Call a managed-package Apex REST endpointapi-apexrest-namespaced-postmedium
33%
Create and update records with sObject Collectionsapi-collections-create-updatemedium
19%
Fetch a record with UI API fields and optionalFieldsapi-uiapi-record-optional-fieldsmedium
0%
Page through a delta query that includes deleted recordsapi-query-paging-queryallmedium
0%
Request a token with the OAuth 2.0 client credentials flowapi-oauth-client-credentialsmedium
97%
Run anonymous Apex through the Tooling APIapi-tooling-execute-anonymousmedium
0%
Upsert an account by external ID with its parentapi-upsert-external-idmedium
44%
Bulk API 2.0 export including deleted recordsapi-bulk2-query-exporthard
17%
Bulk API 2.0 upsert with a parent external IDapi-bulk2-ingest-upserthard
17%
Composite request with an implicit dependencyapi-composite-implicit-dependencyhard
0%
Independent all-or-nothing orders in one callapi-composite-graph-ordershard
8%
Interpret external-ID upsert responsesapi-upsert-response-semanticshard
33%

Want this for your team?

Leo helps Salesforce teams run AI they own: open-weight models on infrastructure you control, tested on your kind of work before you rely on them. A private benchmark of your shortlist is a fixed fee.

Get a private benchmark

Or email leo@azl.au · azl.au