Compatibility matrix
A compatibility matrix by endpoint, model and client version is published before migration is encouraged. A row becomes supported only with a dated measurement on a named pinned id, pool and client version. Until then every row reads unvalidated: starting with OpenAI-style integrations, and a shared client format does not mean every provider feature or model behavior is identical.
Note
Every cell is unvalidated at launch. A row changes only with a dated validation record; the matrix is empty of supported rows until then.
The rows cover the routes HIPS serves or has built: chat completions, Responses, Messages and its token count, models, embeddings, rerank, the image routes, the task API, and files, batches and their webhooks. A route that answers in production is still unvalidated here until its rows pass; its state in production is on the API reference.
Compatibility matrix, generated from its data file, dated per row
| Endpoint | Feature | Client | Client version | Model, pool | Status | Validated |
|---|---|---|---|---|---|---|
| POST /v1/chat/completions | Non-streamed completion | OpenAI Python SDK | Recorded at validation | A pinned catalog model, certified | unvalidated | none |
| POST /v1/chat/completions | Non-streamed completion | OpenAI Node SDK | Recorded at validation | A pinned catalog model, certified | unvalidated | none |
| POST /v1/chat/completions | Non-streamed completion | cURL | none | none | unvalidated | none |
| POST /v1/chat/completions | Non-streamed completion | OpenAI Python SDK | none | A pinned catalog model, community | unvalidated | none |
| POST /v1/chat/completions | Streaming, [DONE] | OpenAI Python SDK, Node SDK | none | both pools | unvalidated | none |
| POST /v1/chat/completions | stream_options.include_usage and the hips object tolerated by the SDK's types | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| POST /v1/chat/completions | Tool calls, non-streamed | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| POST /v1/chat/completions | Tool calls, streamed deltas by index | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| POST /v1/chat/completions | parallel_tool_calls | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| POST /v1/chat/completions | response_format json_object | OpenAI Python SDK, Node SDK | none | both pools | unvalidated | none |
| POST /v1/chat/completions | response_format json_schema, strict: true | OpenAI Python SDK, Node SDK | none | both pools | unvalidated | none |
| POST /v1/chat/completions | Vision image_url parts (data URL, https) | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| POST /v1/chat/completions | logprobs, seed, stop, max_completion_tokens | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| POST /v1/chat/completions | Error envelope parsed: 400, 401, 402, 403, 404, 429 with Retry-After, 503 | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| POST /v1/chat/completions | SDK retries on 429 and 503 and not on 400 | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| POST /v1/chat/completions | Rate-limit headers read | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| POST /v1/chat/completions | Request id read by the SDK | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| POST /v1/chat/completions | Mid-stream error envelope raised by the SDK | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| POST /v1/chat/completions | Requirement headers via default_headers and extra_headers | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| GET /v1/models | models.list() and models.retrieve() with the hips extension tolerated | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| POST /v1/responses | Non-streamed, streamed events, tools, text.format | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| POST /v1/responses | A coding agent's turns through hi codex | Codex | none | none | unvalidated | none |
| POST /v1/messages | Non-streamed and streamed, tool use and tool results, x-api-key, Anthropic's error shape | Anthropic Python SDK, TypeScript SDK | none | none | unvalidated | none |
| POST /v1/messages | A coding agent's turns through hi claude | Claude Code | none | none | unvalidated | none |
| POST /v1/messages/count_tokens | Not served: 404 route_not_available; the client estimates its context itself | Claude Code | none | none | unvalidated | none |
| POST /v1/embeddings | float and base64 encodings, dimensions, the usage object | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| POST /v1/rerank | top_n, return_documents, the results and meta | Cohere's SDKs | none | none | unvalidated | none |
| POST /v1/images/generations | images.generate() with n, size and output_format; b64_json read; hips.withheld tolerated | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| POST /v1/images/edits | images.edit() with one image and with image[]; b64_json read; hips.withheld tolerated | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| POST /v1/files, /v1/batches | Upload, create, retrieve, cancel and list; the output and error files (not offered in production yet) | OpenAI Python SDK, Node SDK | none | none | unvalidated | none |
| Batch webhooks | The four batch events and their Standard Webhooks signatures | OpenAI's webhook verifiers | none | none | unvalidated | none |
| POST /v1/chat/completions | A coding tool's complete tool-call round trip | Each tool separately Not decided yet | none | none | unvalidated | none |
| POST /v1/tasks, /v1/jobs, /v1/uploads | The HIPS SDK conformance suite | HIPS SDK per language Not decided yet | none | none | unvalidated | none |
- Endpoint
- POST /v1/chat/completions
- Feature
- Non-streamed completion
- Client
- OpenAI Python SDK
- Client version
- Recorded at validation
- Model, pool
- A pinned catalog model, certified
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- Non-streamed completion
- Client
- OpenAI Node SDK
- Client version
- Recorded at validation
- Model, pool
- A pinned catalog model, certified
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- Non-streamed completion
- Client
- cURL
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- Non-streamed completion
- Client
- OpenAI Python SDK
- Client version
- none
- Model, pool
- A pinned catalog model, community
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- Streaming, [DONE]
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- both pools
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- stream_options.include_usage and the hips object tolerated by the SDK's types
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- Tool calls, non-streamed
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- Tool calls, streamed deltas by index
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- parallel_tool_calls
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- response_format json_object
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- both pools
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- response_format json_schema, strict: true
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- both pools
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- Vision image_url parts (data URL, https)
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- logprobs, seed, stop, max_completion_tokens
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- Error envelope parsed: 400, 401, 402, 403, 404, 429 with Retry-After, 503
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- SDK retries on 429 and 503 and not on 400
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- Rate-limit headers read
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- Request id read by the SDK
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- Mid-stream error envelope raised by the SDK
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- Requirement headers via default_headers and extra_headers
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- GET /v1/models
- Feature
- models.list() and models.retrieve() with the hips extension tolerated
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/responses
- Feature
- Non-streamed, streamed events, tools, text.format
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/responses
- Feature
- A coding agent's turns through
hi codex - Client
- Codex
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/messages
- Feature
- Non-streamed and streamed, tool use and tool results,
x-api-key, Anthropic's error shape - Client
- Anthropic Python SDK, TypeScript SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/messages
- Feature
- A coding agent's turns through
hi claude - Client
- Claude Code
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/messages/count_tokens
- Feature
- Not served: 404
route_not_available; the client estimates its context itself - Client
- Claude Code
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/embeddings
- Feature
floatandbase64encodings,dimensions, the usage object- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/rerank
- Feature
top_n,return_documents, the results andmeta- Client
- Cohere's SDKs
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/images/generations
- Feature
images.generate()withn,sizeandoutput_format;b64_jsonread;hips.withheldtolerated- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/images/edits
- Feature
images.edit()with one image and withimage[];b64_jsonread;hips.withheldtolerated- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/files, /v1/batches
- Feature
- Upload, create, retrieve, cancel and list; the output and error files (not offered in production yet)
- Client
- OpenAI Python SDK, Node SDK
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- Batch webhooks
- Feature
- The four batch events and their Standard Webhooks signatures
- Client
- OpenAI's webhook verifiers
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/chat/completions
- Feature
- A coding tool's complete tool-call round trip
- Client
- Each tool separately Not decided yet
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
- Endpoint
- POST /v1/tasks, /v1/jobs, /v1/uploads
- Feature
- The HIPS SDK conformance suite
- Client
- HIPS SDK per language Not decided yet
- Client version
- none
- Model, pool
- none
- Status
- unvalidated
- Validated
- none
| Status | Meaning |
|---|---|
| unvalidated | Not yet measured. The default of every row at launch. |
| supported | Measured on the stated client version, pinned id and pool, on the stated date. |
| partial | Measured, with named gaps recorded on the row. |
| unsupported | Measured and refused by design, with the reason recorded. |
- Status
- unvalidated
- Meaning
- Not yet measured. The default of every row at launch.
- Status
- supported
- Meaning
- Measured on the stated client version, pinned id and pool, on the stated date.
- Status
- partial
- Meaning
- Measured, with named gaps recorded on the row.
- Status
- unsupported
- Meaning
- Measured and refused by design, with the reason recorded.
- Error
typevalues are the HIPS list:not_found_erroron an unknown model where OpenAI sendsinvalid_request_error;rate_limit_errorwith the dimension inparamwhere OpenAI usestypefor it. nis 1.storeis refused. The completionidis areq_.- The
hipsextensions inGET /v1/modelsentries and in the final streamed chunk.
Related sheets
- Sheet
- 19 of 20
- Detail
- G
- Updated
- 2026-11-16