Attention: llm.report is no longer actively maintained. The product will still run in the meantime. See Github for more details.

llm.report

Log and Monitor your AI Apps

Know exactly what's happening in your AI app with realtime logging, analytics, usage reports, and alerts.

BACKED BY THE BEST IN THE INDUSTRY

Backed by Vercel AI AcceleratorBacked by Buildspace

Trusted by open-source, enterprise, and more than 3,000 users

Analyze

Advanced OpenAI API Dashboard

Just enter your OpenAI API key, and we fetch your data from the OpenAI API directly to create a dashboard. No need to install anything.

Get Started

Optimize

Log your prompts and completions

Change 1 line in your code and start logging your API requests. Optimize your token usage and start saving money.

Get Started

Minimize

Measure Cost Per User

Analyze your cost per user and adjust your pricing to maximize revenue.

Get Started

Loved by 3,000+ users and companies worldwide

Here's what our users have to say

llm.report is gold πŸ”₯ It lets me get access to a lot of details that OpenAI hides, like the number of tokens I'm sending and receiving. I'm now optimizing and cutting costs like crazy.
are you using text-davinci-003 as your default model? time to switch to gpt-3.5-turbo, which has similar performance but 10x cheaper!! made the switch today, my financial anxiety went away want to visualize your gpt costs? check out llm.report @_buildspace
πŸ”₯ llm.report OSS, instant, realtime LLM observability
llm.report is one of the best examples of a landing page -> fast time to wow, that i've seen. The moment you land, you know instantly what the value prop is, and you're able to interact with a demo report. s/o @dillionverma
Just tried llm.report from @dillionverma It's exactly what we were looking for to use with @assemblymktg Thinking about how to price our AI generation feature and one open question was how much it even cost us and now we know :)
llm.report is gold πŸ”₯ It lets me get access to a lot of details that OpenAI hides, like the number of tokens I'm sending and receiving. I'm now optimizing and cutting costs like crazy.
are you using text-davinci-003 as your default model? time to switch to gpt-3.5-turbo, which has similar performance but 10x cheaper!! made the switch today, my financial anxiety went away want to visualize your gpt costs? check out llm.report @_buildspace
πŸ”₯ llm.report OSS, instant, realtime LLM observability
llm.report is one of the best examples of a landing page -> fast time to wow, that i've seen. The moment you land, you know instantly what the value prop is, and you're able to interact with a demo report. s/o @dillionverma
Just tried llm.report from @dillionverma It's exactly what we were looking for to use with @assemblymktg Thinking about how to price our AI generation feature and one open question was how much it even cost us and now we know :)
Wow, llm.report is fantastic! Definitely recommend checking it out to get some analytics around OpenAI usage. Shout out to @dillionverma for building an awesome product.
Didn't realize that text-davinci-003 was about 50-75% of my Open AI costs, even though I would only send about 5 tokens per request. thanks llm.report
$39.10 out of the $42.11, i.e. ReviewHeroAI.com represented 92% of my total OpenAI usage costs while PlaylistGeniusAI.com represented the other 8%. How was I able to find this out? I used @dillionverma's llm.report tool. Nice work! Feature request -…
love playing around with llm.report by @dillionverma It pulls the data straight from the openai usage api. no wrapping or code changes.
Just tried llm.report, this is absolutely amazing! It helps you keep track your OpenAI APIs spendings as well as log your prompts/outputs. And everything is so simple! Stunning.
This, this we like! Thanks @dillionverma πŸ”₯
Wow, llm.report is fantastic! Definitely recommend checking it out to get some analytics around OpenAI usage. Shout out to @dillionverma for building an awesome product.
Didn't realize that text-davinci-003 was about 50-75% of my Open AI costs, even though I would only send about 5 tokens per request. thanks llm.report
$39.10 out of the $42.11, i.e. ReviewHeroAI.com represented 92% of my total OpenAI usage costs while PlaylistGeniusAI.com represented the other 8%. How was I able to find this out? I used @dillionverma's llm.report tool. Nice work! Feature request -…
love playing around with llm.report by @dillionverma It pulls the data straight from the openai usage api. no wrapping or code changes.
Just tried llm.report, this is absolutely amazing! It helps you keep track your OpenAI APIs spendings as well as log your prompts/outputs. And everything is so simple! Stunning.
This, this we like! Thanks @dillionverma πŸ”₯

Community

Join our community

Join our open-source community and help shape the future of LLM Report.