Docs · Getting started

API reference: every endpoint, with an OpenAPI specification

2026-10-28

Every endpoint TopxAI serves, what each accepts and returns, the two ways to send the key, and the OpenAPI 3.1 file at /openapi.json for API tools.

TopxAI provides OpenAI- and Anthropic-compatible endpoints for the models listed below. Compatibility depends on the model and endpoint: some requests use the provider's native format, while others are converted, and documented fields can be adjusted or removed. A supported endpoint does not guarantee every SDK feature or server-side conversation storage.

Use this page for endpoint selection, authentication and conversion limits. The provider comparison lists field and system-prompt changes; the LangChain guide explains Claude's JSON Schema and tool-choice limits. /openapi.json provides the OpenAPI 3.1 specification for API tools and code generators.

Base URLs

https://ai.topxea.com/v1 for OpenAI-style SDKs and tools; they append /chat/completions, /responses, /images/…, /models.
https://ai.topxea.com for the Anthropic SDK; it appends /v1/messages.

Authentication

A TopxAI key (sk-…) from the API keys page, sent one of two ways:

Authorization: Bearer <key> on every endpoint.
x-api-key: <key> on /v1/messages, the way the Anthropic SDK sends it.

The key's route decides which models it can call; an Auto-route key can call every model.

Endpoints

Endpoint	Format	Models	Notes
`POST /v1/chat/completions`	OpenAI Chat Completions	every text model except Jev	`stream: true` for SSE
`POST /v1/responses`	OpenAI Responses	gpt-6-astra, gpt-6.1-sol, gpt-6.1-terra, grok-4.7, deepseek-flash-Abliterated, kimi-k3.1-pro	what Codex CLI uses
`POST /v1/messages`	Anthropic Messages	every text model except Jev	Claude natively, the other models through a conversion (below); `anthropic-version` optional
`POST /v1/images/generations`	OpenAI Images	gpt-image-2.5-sunburst, gpt-image-2.5-flare	priced per image by the `size` tier; `n` 1 to 128
`POST /v1/images/edits`	OpenAI Images (multipart)	the same two	priced like generation
`POST /v1/videos/generations`	video job	grok-imagine-video-1.5	`duration` 1 to 15 s, `resolution` 480p/720p/1080p; billed on acceptance
`GET /v1/videos/{request_id}`	job status		keep polling on pending, queued, processing, running, in_progress
`GET /v1/videos/{request_id}/content`	the file		after `done`
`POST /v1/systemone`	TypeSafe System One	jev-1.14.0, jev-latest, jev-preview	`model`, `state`, `questions`; images in `assets`, as base64 or multipart parts; also `/v1/system_one`
`GET /v1/models`	OpenAI model list		filtered to the key's route
`GET /api/pricing`	JSON price table		no key needed; the data behind /pricing

For deepseek-flash-Abliterated, native Responses supports stateless requests with full conversation context in input. Server-side conversation persistence is not promised for this deployment. If max_output_tokens is missing, TopxAI fills up to 262,144 tokens, bounded by the context window remaining after the estimated prompt. See the model guide.

kimi-k3.1-pro also accepts stateless Responses requests. Send the full conversation in input; server-side conversation persistence and a complete Codex workflow have not been verified. See the Kimi guide.

Model compatibility and verification

Reviewed on 2026-10-28 against the current catalogue and the linked model guides. This table describes documented support and known limits; it is not a fresh end-to-end test of every model and client version. Use the exact model ID from pricing, and validate the feature your application needs before moving a workload.

Model or family	Interface to use	Features and limits	Setup or evidence
Claude Fable 5.1, Sonnet 5.5, Opus 5.5	Messages for native Claude features; Chat Completions for OpenAI clients	Streaming and tool calls; forced tool choice is rejected. Chat JSON Schema is converted to Claude's supported subset. Native Messages preserves cache breakpoints.	Claude Code setup, structured-output examples and limits
GPT-6 Astra, GPT-6.1 Sol and Terra, Grok 4.7	Chat Completions or Responses; Messages is converted	Streaming is available. Cross-format calls can change reasoning and tool-result representation; the conversion rules below apply. Endpoint support alone does not verify every SDK option.	SDK examples, field changes
Kimi K3.1 Pro	Chat Completions; stateless Responses; converted Messages	Preserve reasoning and tool-call fields across turns. Images use base64 data URIs, not public HTTP image URLs. Responses persistence and a complete Codex workflow remain unverified.	Kimi guide
deepseek-flash-Abliterated	Chat Completions, stateless Responses, converted Messages	Private deployment with text, image input and streaming. Send full conversation context; server-side persistence is not promised. Endpoint support does not establish full Codex workflow acceptance.	Flash guide and request example
GLM-5.3-Abliterated	Chat Completions or converted Messages	Private text-only deployment. Attachments become a text notice. Some JSON Schema string formats become description hints; validate the returned object in your application.	GLM guide
Jev 1.14.0	System One	Typed questions and answers through its own request format; not a Chat Completions, Responses or Messages model.	Jev request examples

For a minimal integration check, start with the SDK request example, then enable streaming or add one tool/schema/image at a time. A successful text reply does not verify those other features. Record the model, route, client version, date and request ID when reporting a result; never publish your API key or private prompt.

Messages requests to other models

Claude models get a /v1/messages request in Anthropic's own format, cache_control breakpoints included; the few fields the relay changes on the way, such as a missing max_tokens set to the model's output limit, are listed in the comparison with the providers' APIs. Without anthropic-version the relay sends Claude 2023-06-01; the converted requests do not carry the header. For the other text models TopxAI converts the request: gpt-6.1-sol and gpt-6.1-terra go to the Responses API; gpt-6-astra, grok-4.7, kimi-k3.1-pro, GLM-5.3-Abliterated and deepseek-flash-Abliterated go to Chat Completions. The answer comes back as a Messages response or event stream either way. Where the converted models differ from Claude:

Without max_tokens, deepseek-flash-Abliterated also gets the model's output limit, cut to what the context window leaves after the estimated prompt. Flash's limit is 262,144 tokens. The other converted models are sent no cap and write up to their provider's default.
The Chat Completions conversion keeps cache_control on the text, image and document blocks of the messages and drops it everywhere else: the system prompt, the tool definitions, tool_use and tool_result blocks. The Responses conversion for gpt-6.1-sol and gpt-6.1-terra drops it everywhere.
On the Chat Completions models a tool_result block keeps its text on the tool message, and its images move to the user message that follows the tool results. A tool_result that holds a document is a 400 there. The Responses conversion passes both on to gpt-6.1-sol and gpt-6.1-terra.
thinking blocks in the conversation history, including those the model's own earlier answers came back with, are dropped before the request reaches gpt-6.1-sol or gpt-6.1-terra; the Chat Completions models take them as reasoning. redacted_thinking blocks are dropped on every converted model. The model continues without the reasoning that was dropped.
grok-4.7 reasons on every request and xAI refuses stop sequences on reasoning models, so stop_sequences on grok-4.7 is a 400 that names the parameter.
GLM-5.3-Abliterated takes text only. On /v1/chat/completions and /v1/messages alike an attachment is replaced by a note (its page has the details).

Status codes

400 the body failed validation, the message names the field; on System One it can also be TypeSafe's own rejection, passed through. 401 key missing or invalid. 402 balance empty. 403 model not sold here, not on this key's route, or excluded by the key's model limits. 404 the provider says the model does not exist, or an unknown video job. 413 a System One body or image over a size limit. 422 TypeSafe's validation detail (System One only). 429 rate limited, Retry-After in seconds. 503 no line can serve the model now; retry. Every error message ends with a request id; quote it to support@topxea.com.

Using the specification

curl -O https://ai.topxea.com/openapi.json

Import the file into Postman or Insomnia to get a request per endpoint with the examples filled in, or feed it to openapi-generator for a typed client. The file's servers entry is the production origin; nothing in it is specific to a key or an account.

Not served

Batch, fine-tuning, files, assistants, vector stores, realtime, audio and embeddings are not relayed; requests to those paths return 404. Keep them on the provider's own key. The comparison with the providers' APIs lists what else stays direct.

Previous: Connect your SDK · Next: Claude Code with TopxAI: ANTHROPIC_BASE_URL and the key

Other languages: 简体中文