
When using Cloudinary's @google font syntax, specify a numeric weight supported by that Google Font. The documented syntax requires numbers such as 400 or 700 instead of keywords such as bold or light. A styling keyword that works with another font route is not proof that it is valid here.
Key Takeaways
- Identify whether the request uses @google, a directly supported font or an uploaded custom file.
- @google weights must be numeric and available in the selected family.
- Successful rendering still needs a visual and wording check.
Keep the font route visible
The official Cloudinary font guide supports Google Fonts through a family name followed by @google. It says names are case-sensitive and must match Google Fonts exactly.
Start by recording the actual family, syntax route, size and requested weight. A regular result from a different route cannot confirm that the intended @google request is configured correctly. Keep the approved wording and an appearance reference while diagnosing, so a fallback is not silently accepted as the desired brand face.
A hypothetical campaign might need a heavier heading in the same Google family. The immediate task is to request a supported weight through the correct grammar, not to flatten a screenshot of the heavier heading.
Put the weight in its documented position
Cloudinary's example uses Roboto at size 90 with weight 600:
l_text:Roboto@google_90_600:Hello%20World
This is an official syntax example, not an executed image request or a recommended size for your design. It shows the size followed by the numeric weight, separated by underscores.
Replace the requested keyword with the appropriate supported numeric weight, keeping family spelling and unrelated styling stable for the first comparison. Do not invent an intermediate weight because the number looks plausible; the documentation requires that the chosen font support it.
Do not confuse this with a missing custom style file
The same source gives a separate rule for uploaded custom fonts: bold and italic use their own uploaded font files. That is a different route from @google's numeric-weight syntax.
The uploaded-family variant guide similarly concerns available style files in another platform, not this Cloudinary parser rule. Follow the requirements of the route actually present in the request.
Accept the rendered face, not only a successful response
Inspect the delivered heading alongside the approved reference at its final size. Read names, punctuation and non-Latin characters, and check that the heavier face has not introduced unwanted line breaks or covered the subject.
The glyph-coverage guide addresses missing characters, while the workspace-font guide concerns asset availability elsewhere. Neither should replace the numeric-weight check. Accept when the actual @google family and supported weight render the complete approved wording; no rendering result or universal font-availability guarantee is claimed.
Related tasks
- Why Are Bold and Italic Disabled for My Uploaded Canva Font?
- Why Does My Print PDF Show the Wrong Font or Empty Character Boxes?
- Why Can I Use a Photoroom Brand Font but My Teammate Can't Find It?
Sources and scope
- Cloudinary: Custom and special fonts: Custom fonts require raw authenticated upload, a lower-case extension, valid path syntax without underscores, and the same product environment. Google font syntax uses @google and numeric font-supported weights instead of bold/light keywords.
Documentation-based guidance from the cited official pages captured on October 25, 2026. Scenarios and acceptance procedures are proposed, not executed tests. No account operation, measured demand, AI-citation frequency, processing benchmark or search-indexing outcome is claimed.
Prepared by Imagild Editorial, the company editorial team responsible for these guides. AI assisted research and drafting. This is documentation-based guidance with proposed examples, not a claim of hands-on testing or an independent product review. Report a correction.
Cover: AI-generated conceptual editorial illustration; not a real application screenshot, observed output or before-and-after test.