Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/Temicide/thcode/llms.txt

Use this file to discover all available pages before exploring further.

T-OCR is the vision-category specialist service in thcode’s AI for Thai catalog. It detects and recognizes Thai text present in an image and returns the recognized text as a string alongside a confidence score and a list of individual words. In the broader thcode workflow, T-OCR results feed directly into the Typhoon reasoning model — after the OCR endpoint produces Thai text, Typhoon can interpret, translate, classify, or incorporate that text into a coding task, bridging the gap between image-embedded Thai content and code.
T-OCR is catalogued in this release. That means it is discoverable and fully inspectable via /tools, but the endpoint mapping, live transport, and contract test are a subsequent integration phase. Invoking T-OCR now returns a stub refusal: "Catalogued — Not available yet".

​
Service Identity

FieldValue
Service IDt-ocr
Upstream nameT-OCR
Thai nameที-โอซีอาร์
Categoryvision
Modalityimage
Endpointhttps://api.aiforthai.in.th/ocr
Support levelcatalogued
Contract ver.1.0.0
Adapter ver.1.0.0

​
Search Terms

The router matches T-OCR against the following terms. You can use any of them in the /tools search command: 
ocr   อ่านตัวอักษร   text recognition   vision

​
Handler: Input Contract

The TocrSpecialistHandler builds a transport request from a binary image artifact. The handler reads the first prepared artifact, validates that it carries bytes (contentKind: 'bytes'), base64-encodes the image data, and assembles the JSON request body.
Request body sent to the OCR endpoint
{
  "image": "<base64-encoded image bytes>",
  "mime": "<artifact media type, e.g. image/png>"
}
The request is dispatched as an authenticated POST to https://api.aiforthai.in.th/ocr with:
  • Authorization: Bearer <key> — the raw API key is resolved in scope only and never appears in Evidence, logs, or the secret-free headersSummary (which shows [redacted]).
  • Content-Type: application/json
If the artifact is absent, empty, or not of kind bytes, the handler throws before a transport call is made.

​
Handler: Output Contract

The TocrSpecialistHandler maps the upstream JSON response into three typed SpecialistFieldValue fields:
Kind: textMapped from the result key in the upstream response. Present when result is a non-empty string. Empty string and null are both treated as not-present and the field is listed in emptyFields.
{ "kind": "text", "value": "สวัสดีชาวโลก", "present": true }
Any key that is absent, null, or of the wrong type produces present: false and is appended to emptyFields. The handler never invents fields that were not in the upstream response.

​
Fixture

The handler ships with a deterministic, built-in (non-user) fixture used for health probes. The fixture is a minimal valid 1×1 red PNG (70 bytes), and the health canary verifies the endpoint accepts it and returns a valid JSON object. The fixture’s expected outputs are: 
EXPECTED_TEXT       = 'สวัสดีชาวโลก'
EXPECTED_CONFIDENCE = 0.95
EXPECTED_WORDS      = ['สวัสดี', 'ชาวโลก']
The canary uses image/png as the media type; any other media type is rejected before the probe is dispatched.

​
What Typhoon Does with T-OCR Results

Once T-OCR returns recognized Thai text, Typhoon receives the recognizedText field as the primary output. From there the reasoning model can:
  • Translate the text into English for an international developer.
  • Use the text as input to a subsequent specialist service (for example, pipe it into Named Entity Recognition).
  • Incorporate the text into code — variable names, comments, string constants, or documentation derived from an image.
  • Describe or summarize the visual content of a scanned document relevant to a coding task.

​
Using /tools with T-OCR

Inspect the T-OCR catalog entry
/tools inspect t-ocr
Search for vision services
/tools search vision
/tools search อ่านตัวอักษร
The inspect output will show invokable: false with reason "Catalogued — Not available yet". Attempting to invoke T-OCR before integration is complete will produce a pre-transport refusal — no network call is made.

​
Phase 1 Role

T-OCR is part of both the Verification Quartet and the Demo Portfolio. In Phase 1, it must pass a live contract test and a complete Typhoon-driven coding scenario before it is considered demo-certified. See ADR 0011 for the full Phase 1 roadmap.

Build docs developers (and LLMs) love

Get started for freeTalk to us