599 posts tagged “llm”
LLM is my command-line tool for running prompts against Large Language Models.
2024
Claude API: PDF support (beta) (via) Claude 3.5 Sonnet now accepts PDFs as attachments:
The new Claude 3.5 Sonnet (
claude-3-5-sonnet-20241022) model now supports PDF input and understands both text and visual content within documents.
I just released llm-claude-3 0.7 with support for the new attachment type (attachments are a very new feature), so now you can do this:
llm install llm-claude-3 --upgrade
llm -m claude-3.5-sonnet 'extract text' -a mydoc.pdf
Visual PDF analysis can also be turned on for the Claude.ai application:
Also new today: Claude now offers a free (albeit rate-limited) token counting API. This addresses a complaint I've had for a while: previously it wasn't possible to accurately estimate the cost of a prompt before sending it to be executed.
docs.jina.ai—the Jina meta-prompt. From Jina AI on Twitter:
curl docs.jina.ai- This is our Meta-Prompt. It allows LLMs to understand our Reader, Embeddings, Reranker, and Classifier APIs for improved codegen. Using the meta-prompt is straightforward. Just copy the prompt into your preferred LLM interface like ChatGPT, Claude, or whatever works for you, add your instructions, and you're set.
The page is served using content negotiation. If you hit it with curl you get plain text, but a browser with text/html in the accept: header gets an explanation along with a convenient copy to clipboard button.
W̶e̶e̶k̶n̶o̶t̶e̶s̶ Monthnotes for October
I try to publish weeknotes at least once every two weeks. It’s been four since the last entry, so I guess this one counts as monthnotes instead.
[... 797 words]Generating Descriptive Weather Reports with LLMs. Drew Breunig produces the first example I've seen in the wild of the new LLM attachments Python API. Drew's Downtown San Francisco Weather Vibes project combines output from a JSON weather API with the latest image from a webcam pointed at downtown San Francisco to produce a weather report "with a style somewhere between Jack Kerouac and J. Peterman".
Here's the Python code that constructs and executes the prompt. The code runs in GitHub Actions.
You can now run prompts against images, audio and video in your terminal using LLM
I released LLM 0.17 last night, the latest version of my combined CLI tool and Python library for interacting with hundreds of different Large Language Models such as GPT-4o, Llama, Claude and Gemini.
[... 1,399 words]python-imgcat (via) I was investigating options for displaying images in a terminal window (for multi-modal logging output of LLM) and I found this neat Python library for displaying images using iTerm 2.
It includes a CLI tool, which means you can run it without installation using uvx like this:
uvx imgcat filename.png
llm-whisper-api. I wanted to run an experiment through the OpenAI Whisper API this morning so I knocked up a very quick plugin for LLM that provides the following interface:
llm install llm-whisper-api
llm whisper-api myfile.mp3 > transcript.txt
It uses the API key that you previously configured using the llm keys set openai command. If you haven't configured one you can pass it as --key XXX instead.
It's a tiny plugin: the source code is here.
Run a prompt to generate and execute jq programs using llm-jq
llm-jq is a brand new plugin for LLM which lets you pipe JSON directly into the llm jq command along with a human-language description of how you’d like to manipulate that JSON and have a jq program generated and executed for you on the fly.
Pelicans on a bicycle. I decided to roll out my own LLM benchmark: how well can different models render an SVG of a pelican riding a bicycle?
I chose that because a) I like pelicans and b) I'm pretty sure there aren't any pelican on a bicycle SVG files floating around (yet) that might have already been sucked into the training data.
My prompt:
Generate an SVG of a pelican riding a bicycle
I've run it through 16 models so far - from OpenAI, Anthropic, Google Gemini and Meta (Llama running on Cerebras), all using my LLM CLI utility. Here's my (Claude assisted) Bash script: generate-svgs.sh
Here's Claude 3.5 Sonnet (2024-06-20) and Claude 3.5 Sonnet (2024-10-22):
Gemini 1.5 Flash 001 and Gemini 1.5 Flash 002:
GPT-4o mini and GPT-4o:
o1-mini and o1-preview:
Cerebras Llama 3.1 70B and Llama 3.1 8B:
And a special mention for Gemini 1.5 Flash 8B:
The rest of them are linked from the README.
llm-cerebras. Cerebras (previously) provides Llama LLMs hosted on custom hardware at ferociously high speeds.
GitHub user irthomasthomas built an LLM plugin that works against their API - which is currently free, albeit with a rate limit of 30 requests per minute for their two models.
llm install llm-cerebras
llm keys set cerebras
# paste key here
llm -m cerebras-llama3.1-70b 'an epic tail of a walrus pirate'
Here's a video showing the speed of that prompt:
The other model is cerebras-llama3.1-8b.
Running prompts against images and PDFs with Google Gemini.
New TIL. I've been experimenting with the Google Gemini APIs for running prompts against images and PDFs (in preparation for finally adding multi-modal support to LLM) - here are my notes on how to send images or PDF files to their API using curl and the base64 -i macOS command.
I figured out the curl incantation first and then got Claude to build me a Bash script that I can execute like this:
prompt-gemini 'extract text' example-handwriting.jpg
Playing with this is really fun. The Gemini models charge less than 1/10th of a cent per image, so it's really inexpensive to try them out.
Un Ministral, des Ministraux (via) Two new models from Mistral: Ministral 3B and Ministral 8B - joining Mixtral, Pixtral, Codestral and Mathstral as weird naming variants on the Mistral theme.
These models set a new frontier in knowledge, commonsense, reasoning, function-calling, and efficiency in the sub-10B category, and can be used or tuned to a variety of uses, from orchestrating agentic workflows to creating specialist task workers. Both models support up to 128k context length (currently 32k on vLLM) and Ministral 8B has a special interleaved sliding-window attention pattern for faster and memory-efficient inference.
Mistral's own benchmarks look impressive, but it's hard to get excited about small on-device models with a non-commercial Mistral Research License (for the 8B) and a contact-us-for-pricing Mistral Commercial License (for the 8B and 3B), given the existence of the extremely high quality Llama 3.1 and 3.2 series of models.
These new models are also available through Mistral's la Plateforme API, priced at $0.1/million tokens (input and output) for the 8B and $0.04/million tokens for the 3B.
The latest release of my llm-mistral plugin for LLM adds aliases for the new models. Previously you could access them like this:
llm mistral refresh # To fetch new models
llm -m mistral/ministral-3b-latest "a poem about pelicans at the park"
llm -m mistral/ministral-8b-latest "a poem about a pelican in french"
With the latest plugin version you can do this:
llm install -U llm-mistral
llm -m ministral-8b "a poem about a pelican in french"
An LLM TDD loop (via) Super neat demo by David Winterbottom, who wrapped my LLM and files-to-prompt tools in a short Bash script that can be fed a file full of Python unit tests and an empty implementation file and will then iterate on that file in a loop until the tests pass.
lm.rs: run inference on Language Models locally on the CPU with Rust (via) Impressive new LLM inference implementation in Rust by Samuel Vitorino. I tried it just now on an M2 Mac with 64GB of RAM and got very snappy performance for this Q8 Llama 3.2 1B, with Activity Monitor reporting 980% CPU usage over 13 threads.
Here's how I compiled the library and ran the model:
cd /tmp
git clone https://github.com/samuel-vitorino/lm.rs
cd lm.rs
RUSTFLAGS="-C target-cpu=native" cargo build --release --bin chat
curl -LO 'https://huggingface.co/samuel-vitorino/Llama-3.2-1B-Instruct-Q8_0-LMRS/resolve/main/tokenizer.bin?download=true'
curl -LO 'https://huggingface.co/samuel-vitorino/Llama-3.2-1B-Instruct-Q8_0-LMRS/resolve/main/llama3.2-1b-it-q80.lmrs?download=true'
./target/release/chat --model llama3.2-1b-it-q80.lmrs --show-metrics
That --show-metrics option added this at the end of a response:
Speed: 26.41 tok/s
It looks like the performance is helped by two key dependencies: wide, which provides data types optimized for SIMD operations and rayon for running parallel iterators across multiple cores (used for matrix multiplication).
(I used LLM and files-to-prompt to help figure this out.)
If we had $1,000,000…. Jacob Kaplan-Moss gave my favorite talk at DjangoCon this year, imagining what the Django Software Foundation could do if it quadrupled its annual income to $1 million and laying out a realistic path for getting there. Jacob suggests leaning more into large donors than increasing our small donor base:
It’s far easier for me to picture convincing eight or ten or fifteen large companies to make large donations than it is to picture increasing our small donor base tenfold. So I think a major donor strategy is probably the most realistic one for us.
So when I talk about major donors, who am I talking about? I’m talking about four major categories: large corporations, high net worth individuals (very wealthy people), grants from governments (e.g. the Sovereign Tech Fund run out of Germany), and private foundations (e.g. the Chan Zuckerberg Initiative, who’s given grants to the PSF in the past).
Also included: a TIL on Turning a conference talk into an annotated presentation. Jacob used my annotated presentation tool to OCR text from images of keynote slides, extracted a Whisper transcript from the YouTube livestream audio and then cleaned that up a little with LLM and Claude 3.5 Sonnet ("Split the content of this transcript up into paragraphs with logical breaks. Add newlines between each paragraph.") before editing and re-writing it all into the final post.