langfuse.com(@langfuse) 's Twitter Profile Photo

😲 We've been impressed with Mirascope AI since we first met them a week ago!
🫳 A pythonic way to build LLM apps just got more even better: a native, decorator-based integration with Langfuse just dropped.

👇 Video Walkthrough below

youtube.com/watch?v=3ikeOa…

account_circle
Mirascope AI(@mirascopeai) 's Twitter Profile Photo

Mirascope now integrates with langfuse.com! Class methods, calls and extractors can now easily be sent to Langfuse for all your trace, eval and metric needs. Check out out walkthrough post here: mirascope.io/post/langfuse-…

account_circle
Maki@Sunwood AI Labs.(@hAru_mAki_ch) 's Twitter Profile Photo

langfuseにデビューした!
✅image: postgres:16.1に変更
✅「docker compose up」で完了!!
github.com/Sunwood-ai-lab…

langfuseにデビューした!
✅image: postgres:16.1に変更
✅「docker compose up」で完了!!
github.com/Sunwood-ai-lab…
account_circle
Maki@Sunwood AI Labs.(@hAru_mAki_ch) 's Twitter Profile Photo

《🎉MOA v0.5.0 リリース~langfuse追加~》
Bedrockの使用量をlangfuseで追跡できるようにしました!
---
ゆくゆくはlangfuseで実験管理もしていきたい!
🔗github.com/Sunwood-ai-lab…

《🎉MOA v0.5.0 リリース~langfuse追加~》
Bedrockの使用量をlangfuseで追跡できるようにしました!
---
ゆくゆくはlangfuseで実験管理もしていきたい!
🔗github.com/Sunwood-ai-lab…
account_circle
Marc Klingen(@MarcKlingen) 's Twitter Profile Photo

Introducing: 💡 Model-based Evaluation in langfuse.com

🫳 Final drop of Langfuse Launch Week: Model-based Evaluations within Langfuse allows for continuous evaluation across various dimensions like hallucinations, toxicity, relevance, and more.

With Langfuse-managed templates and

account_circle
AI Product Engineer(@aipengineer) 's Twitter Profile Photo

For those who have never heard of Langfuse (YC W23), how would Marc Klingen describe langfuse.com to someone who might not be familiar?

The core link is observability and analytics for AI engineers building complex LLM applications.

account_circle
AI Product Engineer(@aipengineer) 's Twitter Profile Photo

Can Marc Klingen walk us through the typical use case scenario of a langfuse.com user?

Yeah, it really depends on the kind of application that you're running.

Most of them are more like LangChain use cases, conversational, or like agents.

account_circle
Diicell(@0xdiicell) 's Twitter Profile Photo

Introducing Dynamite Dogs - built during Burning Heroes's hackathon to generate docs & posts from GitHub repos.

Stack LlamaIndex 🦙, ollama, langfuse.com & Gradio

Claude 3 Haiku prompted by Opus for same quality at ~$0.01/request (40x cheaper).

Code: github.com/diicellman/dyn…

account_circle
Ishaan(@ishaan_jaff) 's Twitter Profile Photo

💥Thrilled to launch groq/llama3 on LiteLLM (YC W23) docs.litellm.ai/docs/providers…

🆕 langfuse.com - Add LLM location logging, and add cache_hit to metadata

🐞 Fix exception when using the 'gemini/gemini-pro-1.5-latest' model

🐞 Fix streaming of completion responses from Anthropic

📝 Log

💥Thrilled to launch groq/llama3 on @LiteLLM  docs.litellm.ai/docs/providers…

🆕 @langfuse  - Add LLM location logging, and add cache_hit to metadata 

🐞 Fix exception when using the 'gemini/gemini-pro-1.5-latest' model

🐞 Fix streaming of completion responses from Anthropic

📝 Log
account_circle
Y Combinator(@ycombinator) 's Twitter Profile Photo

Congrats on the launch, Clemens Rawert, @marcklingen & Max Deichmann!

@Langfuse (YC W23) is building open-source product analytics for LLM apps, helping companies to track and analyze the quality, cost, and latency of LLMs, chains & agents in granular detail.

ycombinator.com/launches/J2s-l…

Congrats on the launch, @rawert, @marcklingen & @maxdeichmann!

@Langfuse (YC W23) is building open-source product analytics for LLM apps, helping companies to track and analyze the quality, cost, and latency of LLMs, chains & agents in granular detail.

ycombinator.com/launches/J2s-l…
account_circle
langfuse.com(@langfuse) 's Twitter Profile Photo

Day 1 support for OpenAI `gpt-4o` including new tokenizer (`o200k_base`) for accurate token counts.
langfuse.com/changelog/2024…

account_circle