Hi There ,
The idea is to reduce the tokens ... so if we use CSV format then comma counted as the token too. And for LLM we can pass any suitable format with optimized tokens .
Hi Jeet, Think about AI agents: they connect to external systems using tools or MCP servers, and their responses typically come back in JSON, not CSV.
Another important point enterprises have invested heavily in REST‑based microservices, and these also return JSON. Because of that, converting JSON into TOON format makes sense.
If your API happens to return CSV instead, that’s fine too you can still use it. I don’t see any issues with that approach.
Hope this helps
For further actions, you may consider blocking this person and/or reporting abuse
We're a place where coders share, stay up-to-date and grow their careers.
It loosk like a CSV.
Why not just use a CSV?
Hi There ,
The idea is to reduce the tokens ... so if we use CSV format then comma counted as the token too. And for LLM we can pass any suitable format with optimized tokens .
Thanks
Sreeni
TOON has commas too, the only difference with tabular data is that TOON's header line has more characters.
TOON:
users[3]{id,name,role,email}:CSV:
id,name,role,emailWhat happens when the data has a comma in it does it keep the quotes to prevent the data comma to be seen as a new column?
Signed in just to like.
Still don't know the benifit!
Hi Jeet, Think about AI agents: they connect to external systems using tools or MCP servers, and their responses typically come back in JSON, not CSV.
Another important point enterprises have invested heavily in REST‑based microservices, and these also return JSON. Because of that, converting JSON into TOON format makes sense.
If your API happens to return CSV instead, that’s fine too you can still use it. I don’t see any issues with that approach.
Hope this helps