Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / predibase/lorax issues and pull requests

#668 - Fix stella embeddings + Integration tests for lorax

Pull Request - State: open - Opened by magdyksaleh 2 days ago

#667 - Set maximum grpc message receive size to 2GiB

Pull Request - State: closed - Opened by tgaddair 2 days ago

#666 - Fix sliding window + compile bug

Pull Request - State: closed - Opened by ajtejankar 4 days ago

#665 - Multi-responses for a single inference

Issue - State: open - Opened by mrcchef 5 days ago

#664 - optimize healthcheck

Pull Request - State: open - Opened by noyoshi 8 days ago

#663 - added metrics docs, updated links in main docs

Pull Request - State: closed - Opened by noyoshi 9 days ago

#662 - Fix absent `fp8_kv` property on llama and qwen models

Pull Request - State: closed - Opened by ajtejankar 11 days ago

#661 - Things started failing after new commit into main

Issue - State: open - Opened by gane5hvarma 11 days ago - 2 comments

#660 - Fix seqlen bug for sliding window models like Mistral v0.1

Pull Request - State: closed - Opened by ajtejankar 11 days ago

#659 - Allow adapter loading for VLMs

Pull Request - State: open - Opened by Infernaught 11 days ago

#658 - Convert to Triton Punica kernels

Pull Request - State: closed - Opened by tgaddair 16 days ago

#657 - Issue: recognizing a base causal language model as an embedding model

Issue - State: open - Opened by veezbo 17 days ago - 1 comment

#656 - Support for Embeddings with XLM-RoBERTa and Adapters

Pull Request - State: closed - Opened by jfhetzer 18 days ago - 1 comment

#655 - Prompt prefix caching for multi-LoRA

Pull Request - State: closed - Opened by tgaddair 18 days ago

#654 - Fix PREDIBASE_API_TOKEN env var being thrown away

Pull Request - State: closed - Opened by joseph-predibase 19 days ago

#653 - Chunked prefill

Pull Request - State: closed - Opened by tgaddair 23 days ago - 2 comments

#652 - Support FP8 KV Cache

Pull Request - State: closed - Opened by ajtejankar 24 days ago - 2 comments

#650 - change runner 2

Pull Request - State: closed - Opened by magdyksaleh 24 days ago

#649 - change runner

Pull Request - State: closed - Opened by magdyksaleh 24 days ago

#648 - Added backwards compatible field to OpenAI json_object API

Pull Request - State: closed - Opened by tgaddair 24 days ago - 1 comment

#647 - arc runner: v2

Pull Request - State: open - Opened by noyoshi 24 days ago

#646 - try using arc runner for build

Pull Request - State: closed - Opened by noyoshi 24 days ago

#645 - Fix compile for qwen-2.5-32b

Pull Request - State: closed - Opened by tgaddair 25 days ago

#644 - Enhance Structured Output Interface

Pull Request - State: closed - Opened by GirinMan 25 days ago - 1 comment

#642 - Otel v2

Pull Request - State: open - Opened by noyoshi 25 days ago

#641 - Fix llava_next for llama 3.2 vision cross attention states

Pull Request - State: closed - Opened by tgaddair 26 days ago

#640 - Look for language model lm head

Pull Request - State: closed - Opened by Infernaught 26 days ago

#639 - Add --disable-sgmv flag

Pull Request - State: closed - Opened by joseph-predibase 26 days ago - 2 comments

#638 - Return n choices for chat completions API

Pull Request - State: closed - Opened by tgaddair 27 days ago

#637 - Phi 3.5 vision (4B model)

Issue - State: open - Opened by CheeseAndMeat about 1 month ago - 2 comments
Labels: enhancement

#636 - Not able to run source code

Issue - State: open - Opened by nirvitarka about 1 month ago

#635 - pass correct stuff to predibase-reporter

Pull Request - State: closed - Opened by magdyksaleh about 1 month ago

#634 - Fix cuda graph tracing without lora ranks

Pull Request - State: closed - Opened by tgaddair about 1 month ago

#633 - Fix FlashInfer when not using prefix caching

Pull Request - State: closed - Opened by tgaddair about 1 month ago

#632 - Fix punica kernel compilation

Pull Request - State: closed - Opened by tgaddair about 1 month ago

#631 - Fix prefix plumbing and BGMV compiler dimensions

Pull Request - State: closed - Opened by tgaddair about 1 month ago

#630 - Added ranks 96 and 128 to BGMV kernel

Pull Request - State: closed - Opened by tgaddair about 1 month ago

#629 - Fix retrace message

Pull Request - State: closed - Opened by tgaddair about 1 month ago

#628 - Fix CUDA graphs for Medusa

Pull Request - State: closed - Opened by tgaddair about 1 month ago

#627 - Fix CUDA graph compilation

Pull Request - State: closed - Opened by tgaddair about 1 month ago

#626 - add label to id this as a lorax image

Pull Request - State: closed - Opened by noyoshi about 1 month ago

#625 - flashinfer backend raises RuntimeError: paged_kv_indices must be a 1D tensor

Issue - State: closed - Opened by baggiponte about 1 month ago - 3 comments
Labels: bug

#624 - Support FP8 KV Cache

Pull Request - State: closed - Opened by ajtejankar about 1 month ago

#623 - Remove LD_PRELOAD from Docker and improve error message

Pull Request - State: closed - Opened by tgaddair about 1 month ago

#622 - Flash mllama

Pull Request - State: closed - Opened by tgaddair about 1 month ago

#621 - support MRL embeddings for qwen2

Pull Request - State: closed - Opened by magdyksaleh about 1 month ago

#620 - Create sync_from_s3.sh to fetch weights from S3 cache

Pull Request - State: closed - Opened by joseph-predibase about 1 month ago - 2 comments

#619 - Added Mllama

Pull Request - State: closed - Opened by tgaddair about 1 month ago

#618 - Add done message to openai endpoints

Pull Request - State: closed - Opened by magdyksaleh about 1 month ago

#617 - Add --predibase-api-token CLI arg

Pull Request - State: closed - Opened by joseph-predibase about 2 months ago

#616 - Append all entries to queue at once for classify_batch

Pull Request - State: closed - Opened by tgaddair about 2 months ago - 1 comment

#615 - add num inputs to metrics

Pull Request - State: closed - Opened by magdyksaleh about 2 months ago

#614 - Fix deps4

Pull Request - State: closed - Opened by magdyksaleh about 2 months ago

#613 - upgrade poetry

Pull Request - State: closed - Opened by magdyksaleh about 2 months ago

#612 - Fix dependencies to address high urgency dependabot alerts

Pull Request - State: closed - Opened by magdyksaleh about 2 months ago

#611 - Performance issues on AWQ and Lora

Issue - State: open - Opened by dumbPy about 2 months ago

#610 - fix flash attention not installed error on cuda 12.6 and driver 560+

Pull Request - State: closed - Opened by noyoshi about 2 months ago - 1 comment

#608 - Fix classify and classify_batch for Python client

Pull Request - State: closed - Opened by tgaddair about 2 months ago

#607 - Issue with loading AWQ quantized Llama 3.1 70B

Issue - State: closed - Opened by dumbPy about 2 months ago - 1 comment

#606 - Running several adapters on the same input

Issue - State: open - Opened by arnaud-secondlayer about 2 months ago - 1 comment

#605 - Can not start lorax from docker

Issue - State: closed - Opened by korlin0110 about 2 months ago - 1 comment

#604 - Added launcher args for preloaded_adapter_source and backend

Pull Request - State: closed - Opened by tgaddair about 2 months ago

#602 - Fix class ner

Pull Request - State: closed - Opened by magdyksaleh about 2 months ago

#600 - Merge weights

Pull Request - State: closed - Opened by magdyksaleh about 2 months ago - 2 comments

#599 - Fail to run server with prefix-caching option

Issue - State: open - Opened by prd-tuong-nguyen 2 months ago - 1 comment

#598 - Speed up NER inference

Pull Request - State: closed - Opened by magdyksaleh 2 months ago - 1 comment

#597 - Support FlashInfer for BERT

Pull Request - State: closed - Opened by tgaddair 2 months ago

#596 - Fix ner entity merging

Pull Request - State: closed - Opened by magdyksaleh 2 months ago

#595 - Flash Attention is not installed?

Issue - State: closed - Opened by ObliviousDonkey 2 months ago - 8 comments

#594 - Fix bert ner

Pull Request - State: closed - Opened by magdyksaleh 2 months ago

#593 - support bge-base-en-v1.5

Pull Request - State: closed - Opened by magdyksaleh 2 months ago

#592 - Issues loading Llama 3.1 8B Instruct

Issue - State: closed - Opened by jonseaberg 2 months ago - 8 comments

#591 - The server is failing to run

Issue - State: open - Opened by u650080 2 months ago

#590 - Add missing configs

Pull Request - State: closed - Opened by magdyksaleh 3 months ago

#589 - Address rust compiler warnings

Pull Request - State: closed - Opened by magdyksaleh 3 months ago

#588 - add new agnostic health endpoint

Pull Request - State: closed - Opened by magdyksaleh 3 months ago

#586 - Add Llava Next (VLM)

Pull Request - State: closed - Opened by tgaddair 3 months ago

#585 - Fix qwen lora

Pull Request - State: closed - Opened by magdyksaleh 3 months ago

#584 - Add prerequisites to readme

Pull Request - State: closed - Opened by csabakecskemeti 3 months ago

#583 - Add "pbase" to `adapter_source` docstrings

Pull Request - State: closed - Opened by alexsherstinsky 3 months ago

#582 - Install flashinfer in Docker

Pull Request - State: closed - Opened by tgaddair 3 months ago

#581 - Add prefix caching

Pull Request - State: closed - Opened by tgaddair 3 months ago - 8 comments

#580 - Added flashinfer support

Pull Request - State: closed - Opened by tgaddair 3 months ago

#578 - Fix outlines compatibility with speculative decoding

Pull Request - State: closed - Opened by tgaddair 3 months ago

#577 - Support classify batch

Pull Request - State: closed - Opened by magdyksaleh 3 months ago

#576 - Adding longrope for serve Phi-3

Pull Request - State: closed - Opened by huytuong010101 3 months ago

#575 - Updated the documentaion about status code

Pull Request - State: open - Opened by Jatintalreja0510 3 months ago - 1 comment

#574 - include prompt and generated tokens as part of logs

Pull Request - State: closed - Opened by noyoshi 3 months ago

#573 - add support for classification in bert

Pull Request - State: closed - Opened by magdyksaleh 3 months ago

#572 - otel fixups

Pull Request - State: open - Opened by noyoshi 3 months ago

#571 - Fix `--compile`

Pull Request - State: closed - Opened by ajtejankar 3 months ago

#570 - Fix adapter mask when using speculative decoding + LM head LoRA

Pull Request - State: closed - Opened by tgaddair 3 months ago