1# Platform Gotchas — Top 10
2 
31. **OSS models require `"trainingType": "globalStandard"`** in the request body — undocumented, and all OSS FT jobs fail without it.
4 
52. **Model catalog `fine_tune` flag is wrong for OSS models** — API returns `fine_tune = false` for all OSS models despite being FT-supported. Hardcode the supported list.
6 
73. **Older SDK versions may fail on `/v1/` project endpoints** — `client.files.create()` throws "API version not supported" with older `openai` package versions. Upgrade to `openai>=1.0` and use the `/v1/` project endpoint (preferred). If you must use an older SDK, fall back to REST API with the non-project `/openai/` endpoint.
8 
94. **ARM "Succeeded" doesn't mean deployment is ready** — `provisioningState: Succeeded` but data plane returns `DeploymentNotReady` indefinitely. Delete and recreate the deployment, then wait ~5 minutes.
10 
115. **OSS FT deployments may fail with InternalServerError** — use the correct provider-specific `model.format` (e.g., `"Mistral AI"` not `"OpenAI"`) and try `capacity=100`.
12 
136. **OSS FT inference hits "Failed to load LoRA" intermittently** — deploy with capacity ≥ 100, use 8+ retries with exponential backoff, and wait 2+ minutes after deployment before first call.
14 
157. **ARM REST and `az cognitiveservices` use different format strings for OSS models** — ARM uses provider names (`"Microsoft"`, `"Meta"`), CLI uses `"OpenAI-OSS"` for all OSS. Mixing them produces HTTP 500.
16 
178. **Content safety false positives on entity extraction data** — PII-dense data (medical records, legal docs, resumes) can trigger "Hate/Fairness" blocks at deployment time. Remove problematic document types.
18 
199. **FT deployments at capacity=1 are severely rate-limited (~1 RPM)** — evaluating 10 samples takes ~10 minutes. Use capacity ≥ 100 for eval workloads and exponential backoff.
20 
2110. **Wrong resource endpoint is a silent killer** — jobs submitted to the wrong Foundry resource succeed via API but don't appear in the portal. Always verify the endpoint matches your Foundry project.
22