mchiang

9 ms·

61.
by mchiang 2y ago
Sorry about that. We are currently uploading the 671B MoE R1 model as well. We needed some extra time to validate it on Ollama.
62.
Stop Paying the OpenAI Tax: The Emerging Open-Source AI Stack(timescale.com)
5 points by mchiang 2y ago|0 comments
63.
Bringing K/V context quantisation to Ollama(smcleod.net)
220 points by mchiang 2y ago|32 comments
64.
Providing Python functions as tools to models in Ollama(ollama.com)
4 points by mchiang 2y ago|0 comments
65.
by mchiang 2y ago
Ollama provides an API endpoint that now supports the ability for an LLM to use tools/functions. Ollama is not a framework itself. Agent Zero already can use Ollama and alternatives to run the LLMs, and this new feature should enable i
66.
by mchiang 2y ago
This is cool. I'd like to give it a try. Press a button, and get GPU access to build apps on.
67.
by mchiang 2y ago
They are available on Hugging Face: https://huggingface.co/collections/google/gemma-2-release-66... Ollama: https://ollama.com/library/gemma2
68.
Meta Releases Llama 3(ai.meta.com)
22 points by mchiang 2y ago|2 comments
69.
by mchiang 3y ago
While the PRs went in slightly earlier, much of the time was spent on testing the integrations, and working with AMD directly to resolve issues. There were issues that we resolved prior to cutting the release, and many reported by the commu
70.
by mchiang 3y ago
Really sorry about this. Do you happen to have logs for us to look into? This is definitely not the way we want
71.
by mchiang 3y ago
There is the qwen 1.5 model from Alibaba team. https://ollama.com/library/qwen ollama run qwen:0.5b ollama run qwen:1.8b ollama run qwen:4b ollama run qwen:7b ollama run qwen:14b ollama run qwen:72b I would only recomm
72.
by mchiang 3y ago
Hi, we’ve been working to support AMD GPUs directly via ROCm. It’s still under development but if you build from source it does work: https://github.com/ollama/ollama/blob/main/docs/development....
73.
by mchiang 3y ago
You can import GGUF, PyTorch or safetensors models into Ollama. I'll caveat that there are current limitations to some model architectures https://github.com/ollama/ollama/blob/main/docs/import.
74.
by mchiang 3y ago
Hi, I'm one of the maintainers on Ollama. We are working on supporting ROCm in the official releases. If you do build from source, it should work (Instructions below): https://github.com/ollama/ollama/blob&#x2
75.
by mchiang 3y ago
It's amazing to see more smaller models being released. This creates opportunities for more developers to run it on their local computers, and makes it easier to fine-tune for specific needs.
76.
Building LLM poowered web apps using client-side technology(blog.langchain.dev)
2 points by mchiang 3y ago|1 comments
77.
by mchiang 3y ago
This model will run on Ollama, llama.cpp, and other tools: ollama run mistral or for llama.cpp, thebloke has uploaded the GGUF models here: https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF/tree/ma... an
78.
by mchiang 3y ago
Hey Dang, sorry about this. Just wanted to clarify that this was a major overhaul to Ollama. In the past, we did not support Linux with GPU support. We needed to change the main architecture to support different GPUs out-of-the-box. We thou
79.
by mchiang 3y ago
Take a look at Databricks' getting a hand from internal team to create the dolly 15k dataset. ( https://www.databricks.com/blog/2023/04/12/dolly-first-open-... ) For training AGI (artificial general
80.
by mchiang 3y ago
disclaimer: I'm one of the maintainers working on Ollama. I would love to hear how you are using Ollama. One of the upcoming releases will involve an official release of Ollama on Linux with CUDA support of the box. From there, we will
81.
by mchiang 3y ago
People have been compiling Ollama to run on Linux. The reason why it's not packaged yet for Linux is due to packaging it with GPU support - at the very least with nvidia support. Almost there!
82.
by mchiang 3y ago
I'll have to download this and try. This is super cool, and thank you for sharing! Ollama does currently support both Apple silicon macs and Intel macs as well.
83.
by mchiang 3y ago
Hey, how are you running this? I just saw someone tweet this and linked here. I ran it, and my result: (I don't know if this code would work) ollama run phind-codellama --verbose "write c code to inject shellcode into remote proce
84.
by mchiang 3y ago
Sorry about that. We are working on building the Windows and Linux versions. May I ask which specific OS you are on?
85.
by mchiang 3y ago
No need to sign up for any service to use Continue with Ollama. This will do the inference all locally. I believe if you want to do the inference using Replicate or Together, you’ll have to sign up for their services.
86.
by mchiang 3y ago
good to see code llama supported through Continue already. Are you seeing good results with code llama yet?
87.
by mchiang 3y ago
still modifying the code completion (foundation / python models) to see what's causing the behavior. Have had some good success with the instruct model: codellama:7b-instruct
88.
by mchiang 3y ago
Ollama supports it already: `ollama run codellama:7b-instruct` https://ollama.ai/blog/run-code-llama-locally More models uploaded as we speak: https://ollama.ai/library/codellama
89.
SeamlessM4T, a Multimodal AI Model for Speech and Text Translation(about.fb.com)
167 points by mchiang 3y ago|35 comments
90.
New llama.cpp format GGUF now merged(github.com)
2 points by mchiang 3y ago|0 comments

More ›