Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
MMMercy2
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
4 ms
·
1.
▲
Fastest JSON Decoding for Local LLMs with Compressed Finite State Machine
(lmsys.org)
2 points
by
MMMercy2
3y ago
|
0 comments
2.
▲
Fast and Expressive LLM Inference with RadixAttention and SGLang
(lmsys.org)
11 points
by
MMMercy2
3y ago
|
0 comments
3.
▲
Databricks picks up MosaicML, an OpenAI competitor, for $1.3B
(techcrunch.com)
3 points
by
MMMercy2
3y ago
|
0 comments
4.
▲
Chatbot Arena Leaderboard: Introducing MT-Bench and Vicuna-33B
(lmsys.org)
8 points
by
MMMercy2
3y ago
|
0 comments
5.
▲
Building a Truly "Open" OpenAI API Server with Open Models Locally
(lmsys.org)
2 points
by
MMMercy2
3y ago
|
0 comments
6.
▲
Chatbot Arena Leaderboard: OpenAI GPT-4 and Anthropic Claude Take the Lead
(twitter.com)
2 points
by
MMMercy2
3y ago
|
0 comments
7.
▲
Chatbot Arena: Benchmarking LLMs in the Wild with Elo Ratings
(lmsys.org)
50 points
by
MMMercy2
3y ago
|
7 comments
8.
▲
Chatbot Arena: side-by-side battles between open LLMs
(chat.lmsys.org)
8 points
by
MMMercy2
3y ago
|
0 comments
9.
▲
by
MMMercy2
3y ago
You can try the smaller 7B version.
10.
▲
by
MMMercy2
3y ago
You can use this command to apply the delta weights. ( https://github.com/lm-sys/FastChat#vicuna-13b ) The delta weights are hosted on huggingface and will be automatically downloaded.
11.
▲
by
MMMercy2
3y ago
They are the parameters of this large language model. There are 13B fp16 numbers.
12.
▲
by
MMMercy2
3y ago
There are certainly some effective language model benchmarks; however, they are not well-suited for evaluating a chat assistant. Some projects employ human evaluation, while this blog post explores an alternative approach based on GPT-4. Bo
13.
▲
by
MMMercy2
3y ago
I am a Vicuna developer. We plan to release the weights once we have addressed all concerns and have a low-resource version of the inference code ready. We released the demo first to get some early feedback on the model.
14.
▲
by
MMMercy2
3y ago
This project fine-tunes LLaMA on ShareGPT and gets competitive performance compared to Google's Bard. https://vicuna.lmsys.org/
15.
▲
by
MMMercy2
4y ago
It can do some basic coding. The code syntax highlighting looks cool. > Do a quick sort in python > Here's an example of a quicksort implementation in Python: def quicksort(array): if len(array) <= 1: return ar