unsloth
vllm
unsloth | vllm | |
---|---|---|
15 | 31 | |
8,974 | 18,931 | |
42.8% | 10.7% | |
9.4 | 9.9 | |
4 days ago | 6 days ago | |
Python | Python | |
Apache License 2.0 | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
unsloth
-
Ask HN: Most efficient way to fine-tune an LLM in 2024?
Gemma 7b is 2.4x faster than HF + FA2.
Check out https://github.com/unslothai/unsloth for full benchmarks!
-
Gemma doesn't suck anymore – 8 bug fixes
Here are the missing links:
* Gemma, a family of open models from Google: https://ai.google.dev/gemma
* Unsloth is a tool/method for training models faster (IIUC): https://github.com/unslothai/unsloth
-
AMD ROCm Software Blogs
Thanks! Again, partnerships over customers. If you're experienced and have the technical chops to make a MI300x sing, we want to work with you. Our model is that we are the capex/opex investor for businesses. As much as I love software, Hot Aisle is more of a hardware business. Running super high end large scale compute is an extreme challenge in itself. We are less interested in building the software side of things and want to foster those who can focus on that side.
https://github.com/unslothai/unsloth/issues/160
https://github.com/search?q=repo%3Apredibase%2Florax+rocm&ty...
https://github.com/sgl-project/sglang/issues/157
https://github.com/casper-hansen/AutoAWQ (supports rocm)
-
Show HN: We got fine-tuning Mistral-7B to not suck
Unsloth’s colab notebooks for fine-tuning Mistral-7B are super easy to use and run fine in just about any colab instance:
https://github.com/unslothai/unsloth
It’s my default now for experimenting and basic training. If I want to get into the weeds with the training, I use axolotl, but 9/10, it’s not really necessary.
-
Mistral 7B Fine-Tune Optimized
If anyone wants to finetune their own Mistral 7b model 2.2x faster and use 62% less memory - give our open source package Unsloth a try! https://github.com/unslothai/unsloth a try! :)
-
Has anyone tried out the ASPEN-Framework for LoRA Fine-Tuning yet and can share their experience?
https://github.com/unslothai/unsloth seems good and more relevant to your aims perhaps but I haven't tried it.
-
Can we discuss MLOps, Deployment, Optimizations, and Speed?
The unsloth project offers some low-level optimizations for Llama et al, and as of today some prelim Mistral work (which I heard is the llama architecture?)
- Show HN: 80% faster, 50% less memory, 0% loss of accuracy Llama finetuning
-
80% faster, 50% less memory, 0% accuracy loss Llama finetuning
This seems to just be a link to the Unsloth Github repo[0], which in turn is the free version of Unsloth Pro/Max[1]. Maybe the link should be changed?
[0]: https://github.com/unslothai/unsloth
- 80% faster, 50% less memory, 0% loss of accuracy Llama finetuning
vllm
-
AI leaderboards are no longer useful. It's time to switch to Pareto curves
I guess the root cause of my claim is that OpenAI won't tell us whether or not GPT-3.5 is an MoE model, and I assumed it wasn't. Since GPT-3.5 is clearly nondeterministic at temp=0, I believed the nondeterminism was due to FPU stuff, and this effect was amplified with GPT-4's MoE. But if GPT-3.5 is also MoE then that's just wrong.
What makes this especially tricky is that small models are truly 100% deterministic at temp=0 because the relative likelihoods are too coarse for FPU issues to be a factor. I had thought 3.5 was big enough that some of its token probabilities were too fine-grained for the FPU. But that's probably wrong.
On the other hand, it's not just GPT, there are currently floating-point difficulties in vllm which significantly affect the determinism of any model run on it: https://github.com/vllm-project/vllm/issues/966 Note that a suggested fix is upcasting to float32. So it's possible that GPT-3.5 is using an especially low-precision float and introducing nondeterminism by saving money on compute costs.
Sadly I do not have the money[1] to actually run a test to falsify any of this. It seems like this would be a good little research project.
[1] Or the time, or the motivation :) But this stuff is expensive.
-
Mistral AI Launches New 8x22B Moe Model
The easiest is to use vllm (https://github.com/vllm-project/vllm) to run it on a Couple of A100's, and you can benchmark this using this library (https://github.com/EleutherAI/lm-evaluation-harness)
- FLaNK AI for 11 March 2024
-
Show HN: We got fine-tuning Mistral-7B to not suck
Great question! scheduling workloads onto GPUs in a way where VRAM is being utilised efficiently was quite the challenge.
What we found was the IO latency for loading model weights into VRAM will kill responsiveness if you don't "re-use" sessions (i.e. where the model weights remain loaded and you run multiple inference sessions over the same loaded weights).
Obviously projects like https://github.com/vllm-project/vllm exist but we needed to build out a scheduler that can run a fleet of GPUs for a matrix of text/image vs inference/finetune sessions.
disclaimer: I work on Helix
-
Mistral CEO confirms 'leak' of new open source AI model nearing GPT4 performance
FYI, vLLM also just added experiment multi-lora support: https://github.com/vllm-project/vllm/releases/tag/v0.3.0
Also check out the new prefix caching, I see huge potential for batch processing purposes there!
- VLLM Sacrifices Accuracy for Speed
- Easy, fast, and cheap LLM serving for everyone
- vllm
- Mixtral Expert Parallelism
- Mixtral 8x7B Support
What are some alternatives?
DeepSpeed - DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
TensorRT - NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This repository contains the open source components of TensorRT.
llama.cpp - LLM inference in C/C++
CTranslate2 - Fast inference engine for Transformer models
nanoChatGPT - nanogpt turned into a chat model
lmdeploy - LMDeploy is a toolkit for compressing, deploying, and serving LLMs.
gpt-fast - Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.
Llama-2-Onnx
transformers - 🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
tritony - Tiny configuration for Triton Inference Server
accelerate - 🚀 A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (including fp8), and easy-to-configure FSDP and DeepSpeed support
faster-whisper - Faster Whisper transcription with CTranslate2