Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / PygmalionAI/aphrodite-engine issues and pull requests
#882 - chore: consolidate environment variables within one file
Pull Request -
State: closed - Opened by AlpinDale 2 months ago
#881 - api: better startup failure UX
Pull Request -
State: closed - Opened by AlpinDale 2 months ago
#880 - async: disable multi-step scheduling for sync engine
Pull Request -
State: closed - Opened by AlpinDale 2 months ago
#879 - build: pass `PYTHONPATH` from setup.py to cmake
Pull Request -
State: closed - Opened by AlpinDale 2 months ago
#878 - api: fix crashes under very high loads
Pull Request -
State: closed - Opened by AlpinDale 2 months ago
#877 - fix: `add_generation_template` -> `add_generation_prompt` in llm
Pull Request -
State: closed - Opened by AlpinDale 2 months ago
#876 - Update README.md
Pull Request -
State: closed - Opened by NoahBPeterson 2 months ago
#875 - sampler: pad dry sequence breakers tensor
Pull Request -
State: closed - Opened by AlpinDale 2 months ago
- 1 comment
#874 - spec decoding: set the draft model ctxlen to target model
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#873 - cpu: fix `mm_limits` initialization
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#872 - async: avoid premature exit in the async generator
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#871 - feat: add support for chunked prefill + prefix caching
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#870 - lora: add scaling factor support for LoRA at runtime
Pull Request -
State: open - Opened by AlpinDale 3 months ago
#869 - fix: temp_last warning being repeated for every output token
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#868 - Rewrite DRY sampler to be a lot faster
Pull Request -
State: closed - Opened by 50h100a 3 months ago
- 1 comment
#867 - feat: AWQ quantization for InternVL
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#866 - server: log the process occupying our port
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#866 - server: log the process occupying our port
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#865 - executor: pipe `worker_class_fn` arg in executor
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#865 - executor: pipe `worker_class_fn` arg in executor
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#864 - xpu: disable punica kernels for XPU
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#864 - xpu: disable punica kernels for XPU
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#863 - attention: add `AttentionState` abstraction
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#863 - attention: add `AttentionState` abstraction
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#862 - build: add jinja2 to requirements file
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#862 - build: add jinja2 to requirements file
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#861 - xpu: refactor XPU worker & executor
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#860 - dry: only apply dry to sequences that request it
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#859 - ci: bump aphrodite version to 0.6.4.post1
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#858 - sampler: allow parsing sampler order using strings
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#857 - Update sampler.py
Pull Request -
State: closed - Opened by gitzaidi 3 months ago
#856 - sampler: optimize DRY performance using z-algorithm
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
- 4 comments
#855 - sampler: add range parameter for DRY
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#854 - [Bug]: ModuleNotFoundError: No module named 'ray'
Issue -
State: open - Opened by gizbo 3 months ago
- 4 comments
Labels: bug
#853 - [Bug]: Generation sometimes slows to a crawl for all requests when there is a DRY sampler request
Issue -
State: open - Opened by Nero10578 3 months ago
- 9 comments
Labels: bug
#852 - sampler: fix DRY concurrency issue
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#851 - add linux arm64/aarch64/GH200 installation tips
Pull Request -
State: closed - Opened by qpwo 3 months ago
- 1 comment
#850 - fix: outlines import errors
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#849 - DRY Fix: Add output_tokens to sampler
Pull Request -
State: closed - Opened by selalipop 3 months ago
- 3 comments
#848 - [Bug]: DRY has no effect on output
Issue -
State: closed - Opened by discordianbelle 3 months ago
- 4 comments
Labels: bug
#847 - [Bug]: loading a GPTQ-INT4 model on windows with a P40
Issue -
State: open - Opened by sorasoras 3 months ago
Labels: bug
#846 - [Bug]: Crashes whole server when using DRY sampler
Issue -
State: closed - Opened by Nero10578 3 months ago
- 1 comment
Labels: bug
#845 - ci: bump version to 0.6.4
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#844 - chore: bump mistral_common to 1.5.0
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#843 - fix: disable awq_marlin override for awq models
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#842 - feat: Machete Kernels for Hopper GPUs
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#842 - feat: Machete Kernels for Hopper GPUs
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#841 - fix: latency and serving benchmarks
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#841 - fix: latency and serving benchmarks
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#840 - chore: refactor executor classes for easier inheritance
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#840 - chore: refactor executor classes for easier inheritance
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#839 - fix: hidden states handling in batch expansion for spec decoding
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#839 - fix: hidden states handling in batch expansion for spec decoding
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#838 - passthrough arg from api to model.forward
Pull Request -
State: open - Opened by qpwo 3 months ago
- 14 comments
#838 - passthrough arg from api to model.forward
Pull Request -
State: closed - Opened by qpwo 3 months ago
- 17 comments
#837 - feat: add sampler_priorty
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
- 1 comment
#836 - [Feature]: pass-through parameter from request to model.forward (already implemented)
Issue -
State: open - Opened by qpwo 3 months ago
- 1 comment
#836 - [Feature]: pass-through parameter from request to model.forward (already implemented)
Issue -
State: open - Opened by qpwo 3 months ago
- 1 comment
#835 - feat: Classifer-Free Guidance (take 2)
Pull Request -
State: open - Opened by AlpinDale 3 months ago
- 1 comment
#835 - feat: Classifer-Free Guidance (take 2)
Pull Request -
State: open - Opened by AlpinDale 3 months ago
- 1 comment
#834 - feat: add skew sampling
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#834 - feat: add skew sampling
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#833 - WIP: banned strings
Pull Request -
State: open - Opened by AlpinDale 3 months ago
#833 - WIP: banned strings
Pull Request -
State: open - Opened by AlpinDale 3 months ago
#832 - feat: add no_repeat_ngram sampler
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#832 - feat: add no_repeat_ngram sampler
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#831 - feat: multi-step scheduling
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#830 - fix: unbound tokenizer error
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#829 - feat: add metrics for prefix cache hit rate
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#828 - feat: add cuda sampling kernels for top_k and top_p
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#827 - feat: Add DRY (Do not Repeat Yourself) sampling
Pull Request -
State: closed - Opened by selalipop 3 months ago
- 13 comments
#826 - fix: sampler test with new transformers version
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#825 - feat: implement top-nsigma sampling method
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
- 7 comments
#824 - SPMD optimizations
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#823 - feat: support chunked prefill with LoRA
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#822 - feat: add chat method for LLM class
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#821 - fix: tokenization api test
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#820 - [Tracker]: Passing all unit tests
Issue -
State: open - Opened by AlpinDale 3 months ago
Labels: help wanted
#819 - build(deps): bump cross-spawn from 7.0.3 to 7.0.5 in /docs
Pull Request -
State: open - Opened by dependabot[bot] 3 months ago
Labels: dependencies
#818 - fix: --max-seq-len-to-capture arg
Pull Request -
State: closed - Opened by AlpinDale 3 months ago
#817 - fix: ROCm build
Pull Request -
State: closed - Opened by Naomiusearch 3 months ago
- 3 comments
#816 - [Bug]: Argument --max-seq_len-to-capture not recognized
Issue -
State: closed - Opened by Nero10578 3 months ago
- 1 comment
Labels: bug
#815 - [Installation]: Cannot find CUDA_TOOLKIT_ROOT_DIR while trying to build for ROCm
Issue -
State: open - Opened by RuntimeRacer 3 months ago
- 1 comment
#814 - fix: temperature issues
Pull Request -
State: closed - Opened by 50h100a 3 months ago
#813 - Mask dynatemp using min/max, rather than exp
Pull Request -
State: closed - Opened by 50h100a 3 months ago
#812 - [Usage]: Aphrodite Engine: KV Cache Context Length Issue with Quantized Models
Issue -
State: closed - Opened by murtaza-nasir 3 months ago
- 1 comment
#811 - feat: add Tencent Hunyuan model support
Pull Request -
State: open - Opened by AlpinDale 3 months ago
#810 - [Bug]: v0.6.3(.post1?) regression
Issue -
State: open - Opened by dirkson 3 months ago
- 1 comment
Labels: bug
#809 - [Bug]: 0.6.3.post1 regression: RuntimeError during mem profiling on Mistral Large AWQ with `-q awq_marlin`
Issue -
State: open - Opened by khanonnie 4 months ago
- 2 comments
Labels: bug
#808 - feat: update to serviceinfo v0.2
Pull Request -
State: closed - Opened by AlpinDale 4 months ago
#807 - feat: add serviceinfo endpoint
Pull Request -
State: closed - Opened by AlpinDale 4 months ago
#806 - [Misc]: log input and output
Issue -
State: closed - Opened by Eve-146T 4 months ago
- 1 comment
#806 - [Misc]: log input and output
Issue -
State: closed - Opened by Eve-146T 4 months ago
- 1 comment
#805 - frontend: add an `ai-plugin.json` route
Pull Request -
State: closed - Opened by AlpinDale 4 months ago
- 1 comment
#804 - [Bug]: .\gguf_to_torch.py broken along with direct load GGUF
Issue -
State: open - Opened by sorasoras 4 months ago
- 2 comments
Labels: bug
#804 - [Bug]: .\gguf_to_torch.py broken along with direct load GGUF
Issue -
State: open - Opened by sorasoras 4 months ago
- 2 comments
Labels: bug
#803 - frontend: enable kobold api by default
Pull Request -
State: closed - Opened by AlpinDale 4 months ago
#803 - frontend: enable kobold api by default
Pull Request -
State: closed - Opened by AlpinDale 4 months ago
#802 - [Bug]: The documentation page is down and empty
Issue -
State: open - Opened by puppetm4st3r 4 months ago
- 5 comments
Labels: bug
#802 - [Bug]: The documentation page is down and empty
Issue -
State: closed - Opened by puppetm4st3r 4 months ago
- 5 comments
Labels: bug