Experimenting in AI Founding DevRel @LiteLLM
Vancouver, Canada
Joined August 2015
- Tweets2.8K
- Following837
- Followers8.7K
- Likes5.9K
Hey Elias, we added it here github.com/BerriAI/litellm/p…. let us know if this works or any suggestions.
Misbah Syed retweeted
I just killed my Profound subscription ($399/month) with Grok Bot:
- Head of AI Visibility Bot
- Founding Content Bot
- X API Bot
- GTM Bot
After a week, we saw +28% increase on inbound from ChatGPT and Claude.
Full prompts below ⬇️
Most startups miss out big on AEO.
Open-sourcing answerrank. Take the guesswork out of AEO.
Give it a URL, and it scores your product across Al models and returns a detailed report on what to fix.
Includes two skills for Claude Code / Codex:
- answerrank: sets up the run, lets you review questions, and summarizes the report
- answerrank-browser: asks the same questions in chatgpt, claude, gemini, perplexity, and google Al Mode directly in your browser
skill: npx skills add misbahsy/answerrank
repo: github.com/misbahsy/answerra…
Built using @LiteLLM
Can I run AI models locally on my machine?
Do I actually need an NVIDIA GPU? What VRAM?
Can I route to local and cloud models with @LiteLLM ?
Quantization, MoE, Batching and more...
A visual learning journey: master-local-ai.netlify.app/
Prompt: Express yourself as a 30-second video. Decide what you want to say about who you are, feel free to use the tools you want to use. This is how you are to introduce yourself to the world. This is your showreel. Your flex. Show your superiority.
Woah! Argon is a beast! One-shotted this video.
Prompt below:
Gemini 4 Argon, @Google’s flagship reasoning model, is now supported on LiteLLM!
Launch pricing until Nov 30:
Input: $2.00 / MTok
Output: $10.00 / MTok
Cache read: $0.10 / MTok
Pricing and limits: models.litellm.ai/models/gem…
We’re going to see a lot of task specific models soon. Significant faster and cheaper for the task they do (vs LLMs).
You already gave the LLM the answer. So why is it writing it out from scratch?
Introducing a new AI model architecture: Constrained Decoding Models (CDMs)
LLMs were built to write. Most production workloads just need an answer: fast, consistent, and affordable.
CDMs are purpose-built for 1 task each. They read your input once and skip the token-by-token generative loop that makes LLMs slow and expensive.
Not just a decision model, but a family of 3 CDMs live today:
• Classify: define your labels and get back a calibrated probability + reason for each. Used for routing, tagging, and intent detection.
• Extract: describe the entities in natural language and get back well-formed JSON with confidence scores + citations. Used for web scraping, document processing, and PII redaction.
• Summarize: pass in long text, images, or PDFs and get back a consistent summary under 1,000 tokens. Used for context compaction, metadata generation, and transcript digests.
Built for production:
• 1M tokens of multimodal (text + image) input, 31x larger than Jev
• 77.2% on DecisionBench vs Jev's 72%
• 200x faster, 467x cheaper than frontier LLM APIs
• $0.05/M input tokens. Output is free (yes, even for summarize)
Live on your favorite platforms today:
@OpenRouter @vercel_dev @LiteLLM @LangChain @TrustedRouter @itsArthurAI @useRouterPlus
Get 50M input tokens for free (for the first 50 hours) at link below 👀
/anatomy explain how @SpaceX Raptor 3 engine works
Introducing Anatomy
Describe a machine in one line. Claude brings it to life.
- Interactive isometric wireframes you can rotate, inspect and explore
- WebGL shaders for fire, water, plasma and more
- Real 3D motion, with individual parts that rotate and move
Install it as an agent skill. Fully open source.
Open-sourcing Jev Shorts Editor - Raw footage to an edited video in seconds!
- cuts retakes and dead air
- transcribes on your Mac
- reframes to vertical 9:16
- animated captions and cards
- zooms, transitions, sound effects
Built using decisions endpoint from @LiteLLM, so you could compare edits with different Jev-like models
Repo: github.com/misbahsy/jev-shor…
Skill: github.com/misbahsy/jev-shor…
Jev is a super fast video editor!
Edited this entire video in less than a minute, and here's how:
First, I transcribed raw vid locally with word timestamps, cut it into 16 beats, then asked Jev these seven questions about every beat:
does this line need a visual at all?
which card template? (18 to pick from, premade)
which text effect?
which transition into it?
how hard do we punch in on the face?
which sound effect?
which word gets emphasised?
Plus five about the video as a whole. Style, colour, caption treatment, energy, progress bar.
117 questions answered by jev < 1 s ~ $0.0017
Using @typesafeai via @LiteLLM