Skip to main content

Inceptron Provider Guide

Inceptron is a Tier-2 catalog provider: an OpenAI-compatible endpoint described by one JSON file (src/lib/providers/catalog/inceptron.json) rather than hand-written code.

Verification status: this entry is docs- and roster-verified, not yet live-verified. The fields below come from Inceptron's own public pages, an unauthenticated GET /v1/models call and an unauthenticated GET /v1/chat/completions call — no account was created and no API key was used to build it. evidence.liveMatrix is null until someone runs the live capability matrix with a real key (see Verification status below).

Key Facts​

  • Provider id: inceptron
  • Protocol: OpenAI-compatible (/chat/completions) — the Chat Completions API Reference says "Our API is compliant with OpenAI's Chat Completions API" (docs)
  • Base URL: https://api.inceptron.io/v1
  • Default model: zai-org/GLM-5.3
  • Models in catalog: 6 (the full 6-model live roster)
  • Streaming: supported — the Chat Completions API Reference documents the stream parameter and a streaming response, and there is a Handling Streaming Responses guide
  • Tool calling: true — the Chat Completions API Reference and the Tool Calling guide describe OpenAI-compatible tool calling with tools and tool_choice, with worked examples on moonshotai/Kimi-K2.6. The models page tags each of the six model cards tool-calling
  • Tools while streaming: true — the Chat Completions API Reference has a "Streaming Tool Call Response" section and the Tool Calling guide has a "Handle Streamed Tool Calls" section
  • Structured output: false on this entry. The Chat Completions API Reference request table lists model, messages, temperature, max_tokens, stream, top_p, n, stop, tools and tool_choice; the roster's supported_features field lists json_mode and structured_outputs for the six models. Structured output was not exercised live
  • Structured output + tools together: not declared (false) — no combined probe was run without credentials
  • Embeddings: not declared (false)
  • Thinking: not declared (false). The models page tags each of the six model cards reasoning, and the Kilo Code guide configures reasoningEffort variants low, high and max for zai-org/GLM-5.3
  • Billing: recorded as no-free-tier because the catalog schema has no unknown value — a schema placeholder, not a vendor statement
  • Key format: none declared

Quick Start​

1. Get an API key​

  1. Visit: https://console.inceptron.io and sign in or sign up — the sign-in page shows Continue with Google, Continue with GitHub and Sign in with email, and links to https://console.inceptron.io/auth/signup
  2. Create an API key — the Introduction page says: "After signing up, navigate to the account section of the dashboard to create an API key." (https://docs.inceptron.io/introduction)
  3. Set INCEPTRON_API_KEY in your .env file (the docs' examples read INCEPTRON_API_KEY)

The Authentication guide (docs) says:

You can find your API key in your Inceptron account dashboard under the "API Keys" section.

2. Configure​

export INCEPTRON_API_KEY=your-api-key
export INCEPTRON_MODEL=zai-org/GLM-5.3   # optional — overrides the default model
export INCEPTRON_BASE_URL=https://api.inceptron.io/v1   # optional — proxy or self-hosted gateway

3. Use it​

import { NeuroLink } from "@juspay/neurolink";

const neurolink = new NeuroLink();

const result = await neurolink.generate({
  input: { text: "Explain context windows in one paragraph." },
  provider: "inceptron",
  model: "zai-org/GLM-5.3",
});

console.log(result.content);
# CLI
npx @juspay/neurolink generate "Hello" --provider inceptron

Per-request credentials also work:

await neurolink.generate({
  input: { text: "Hello" },
  provider: "inceptron",
  credentials: { inceptron: { apiKey: process.env.INCEPTRON_API_KEY } },
});

Models​

ModelContextVision$/M in · out (cache read)Notes
zai-org/GLM-5.3 ⭐1,048,576no$1.40 / $4.40 ($0.26)NeuroLink default. The model in the docs' Get Started, Introduction, Chat Completions, Error Handling and Streaming code examples and in the Kilo Code and OpenCode config. Tags: text, code, tool-calling, reasoning.
moonshotai/Kimi-K2.6262,144yes$0.73 / $3.50 ($0.25)NeuroLink fallback. The model in the docs' tool-calling examples. Tags: multimodal, tool-calling, reasoning. models.visionModel.
deepseek-ai/DeepSeek-V4-Flash-07311,048,576no$0.13 / $0.28 ($0.03)NeuroLink fallback. Tags: text, code, tool-calling, reasoning.
zai-org/GLM-5.21,048,576no$1.20 / $4.20 ($0.26)NeuroLink fallback. Tags: text, code, tool-calling, reasoning.
moonshotai/Kimi-K2.7-Code262,144yes$0.75 / $3.50 ($0.20)Tags: multimodal, tool-calling, reasoning.
zai-org/GLM-5.3-Flash1,048,576no$0.15 / $0.50 ($0.07)The vendor homepage shows the text "Try GLM 5.3 Flash". The models page shows "1.3M context"; the roster context_length is 1048576 and contextWindow uses that figure. Tags: text, code, tool-calling, reasoning.

Prices, the context labels ("1M context", "262K context") and the tags come from the models page, https://www.inceptron.io/models, opened 2026-09-29. The exact context figures 1,048,576 and 262,144 are the context_length values in an unauthenticated GET https://api.inceptron.io/v1/models call made the same day; for zai-org/GLM-5.3 the Kilo Code guide also states "GLM-5.3 supports a 1M-token context window (1,048,576 tokens)" (docs). The pricing field in the roster response carries different values from the models page for each of the 6 models; this entry uses the models page.

The roster's max_output_length is 1048576 for zai-org/GLM-5.3, deepseek-ai/DeepSeek-V4-Flash-0731, zai-org/GLM-5.2 and zai-org/GLM-5.3-Flash, and 262144 for moonshotai/Kimi-K2.6 and moonshotai/Kimi-K2.7-Code. The Kilo Code guide's config block for zai-org/GLM-5.3 sets limit to context 1048576 and output 1048576. No per-model maxOutputTokens is set in the catalog entry.

models.defaultContextWindow (128,000) and models.defaultMaxOutputTokens (16,384) are NeuroLink placeholders for model ids outside the catalog, not vendor figures. Each model's status field holds production, a placeholder for a value the catalog schema requires; it is not a vendor statement.

The docs' Models API Reference lists 18 ids: deepseek-ai/DeepSeek-V4-Flash-0731, zai-org/GLM-5.3, moonshotai/Kimi-K2.7-Code, moonshotai/Kimi-K2.6 and zai-org/GLM-5.2 with Deployment Mode Serverless, and 13 more with Deployment Mode Custom. The 13 Custom ids are not in the live roster and not in this catalog.

Vision: moonshotai/Kimi-K2.6 and moonshotai/Kimi-K2.7-Code are vision: true from the models page tag multimodal and the roster's input_modalities of ["text","image"]; no image request was sent. The other four models have input_modalities of ["text"] on the roster and are vision: false.

Fallback order when the default is unavailable: moonshotai/Kimi-K2.6 → deepseek-ai/DeepSeek-V4-Flash-0731 → zai-org/GLM-5.2. The runtime fallback model name the loader derives (fallbacks[1]) is deepseek-ai/DeepSeek-V4-Flash-0731.

Verification status​

Tier-2 onboarding requires evidence before a provider is accepted, and pnpm run verify:provider-onboarding gates it in CI. This is what the catalog currently records for Inceptron — docs- and roster-verified, not live-verified:

ProbeResult
Rosterunauthenticated GET /v1/models, HTTP 200, 6 models, 2026-09-29 — no key used
Auth-failure shapeunauthenticated GET https://api.inceptron.io/v1/chat/completions, HTTP 401, body {"error":{"code":"unauthorized","message":"Missing or invalid authorization header","type":"invalid_request_error"}}, 2026-09-29 — no key used. errorRules match 401 (probe and docs) and 429 (docs)
Tools / structured outputtool calling and streamed tool calls are documented on the Chat Completions API Reference and the Tool Calling guide; neither tools nor structured output was exercised live, and no combined tools+schema request was sent — structuredOutputWithTools stays false
Live capability sweepnot run. evidence.liveMatrix is null. Before treating this provider as production-ready, run npx tsx test/continuous-test-suite-provider-matrix.ts --provider=inceptron with a real key and record the result.

Do not treat this entry as equivalent to a live-verified Tier-2 provider (e.g. FriendliAI, Novita AI) until that live matrix has been run and evidence.liveMatrix is filled in.

Troubleshooting​

The Error Handling page (https://docs.inceptron.io/Guides/error-handling) lists these HTTP status codes, quoted as it shows them:

StatusVendor's description
400Your request was malformed. This could be due to missing required parameters or invalid JSON.
401Your API key is missing, invalid, or disabled.
429You have exceeded your rate limit.
500Something went wrong on our end. Please wait and try again.

Pages opened for this entry​

Opened on 2026-09-29 with unauthenticated GET requests:

See also​