The Token Company (YC W26)
19 posts
The Token Company (YC W26)
@thetokenco
Compress LLM input tokens to improve accuracy and save on costs
San Francisco, US Katılım Ocak 2026
3 Takip Edilen369 Takipçiler
Try out the before and after at app.thetokencompany.com
English
Evaluating compressed prompts just got easier.
You can now compare LLM outputs before and after compression in our Compression sandbox to directly evaluate your use case.
English
"Grok make the lobster wear a cape with our logo on it"
Grok:
Harj Taggar@harjtaggar
Does anyone know what’s going on with the lobster on Wall Street lol?
English
The Token Company (YC W26) retweetledi
An AI application client of ours has saved $10k worth of LLM tokens within a week of integrating.
Their approach is just to compress ~3% of inputs removing the absolutely most useless tokens achieving full accuracy with extra cost savings.
With their volume, all of this adds up. Had to 10X rate limits. Now back to work.
English
The Token Company (YC W26) retweetledi
i built a plugin that saves you hundreds of dollars in @opencode
by using the @thetokenco compression model, it shrinks the amount of tokens in your input query before hitting the models
the queries are faster and cheaper, all while maintaining output quality
Y Combinator@ycombinator
The Token Company (@thetokenco) builds LLM input optimization to lower costs, reduce latency, and improve accuracy. Congrats on the launch, @OtsoVeistera! ycombinator.com/launches/Pb3-t…
English
The Token Company (YC W26) retweetledi
Link to the article: thetokencompany.com/benchmarks/fin…
English
Compression absolutely crushed SEC filings
Does anyone actually read these? I don't know, but we let our bear-1.2 LLM compression model do that on FinanceBench (150 real SEC filing questions)
Bro got up to 84.7% accuracy (vs 82% baseline) with 20% token reduction. Saves costs, boosts speed, crushes financial analysis
Removing bloat makes the model perform better
English
The Token Company (YC W26) retweetledi
English
The Token Company (YC W26) retweetledi
The Token Company (@thetokenco) builds LLM input optimization to lower costs, reduce latency, and improve accuracy.
Congrats on the launch, @OtsoVeistera!
ycombinator.com/launches/Pb3-t…
English
The Token Company (YC W26) retweetledi
The Token Company (YC W26) retweetledi
Incredibly excited about this! 🔥🔥🔥
otso veistera@OtsoVeistera
You're wasting half your context window. We’re launching @thetokenco (YC W26) today. We compress LLM inputs before they reach the model. Fewer tokens, lower cost, faster inference. Models also perform better. In customer case studies we’ve seen a +5% lift in user purchases due to higher preference for outputs from compressed prompts. The API is live. Link in the comments
English
The Token Company (YC W26) retweetledi
we’ve launched @thetokenco!
otso veistera@OtsoVeistera
You're wasting half your context window. We’re launching @thetokenco (YC W26) today. We compress LLM inputs before they reach the model. Fewer tokens, lower cost, faster inference. Models also perform better. In customer case studies we’ve seen a +5% lift in user purchases due to higher preference for outputs from compressed prompts. The API is live. Link in the comments
English
The Token Company (YC W26) retweetledi
seems like all our residents are launching today
congratulations @OtsoVeistera & @rasmus_up 🎉
otso veistera@OtsoVeistera
You're wasting half your context window. We’re launching @thetokenco (YC W26) today. We compress LLM inputs before they reach the model. Fewer tokens, lower cost, faster inference. Models also perform better. In customer case studies we’ve seen a +5% lift in user purchases due to higher preference for outputs from compressed prompts. The API is live. Link in the comments
English
The Token Company (YC W26) retweetledi
You're wasting half your context window.
We’re launching @thetokenco (YC W26) today.
We compress LLM inputs before they reach the model.
Fewer tokens, lower cost, faster inference.
Models also perform better. In customer case studies we’ve seen a +5% lift in user purchases due to higher preference for outputs from compressed prompts.
The API is live. Link in the comments
English