rustformers/llm issues and pull requests

#184 - nix: add flake

Pull Request - State: closed - Opened by cmoog over 1 year ago - 2 comments

#183 - Update upstream GGML dependency

Pull Request - State: closed - Opened by danforbes over 1 year ago

#182 - Docs and refactors for Model stuff

Pull Request - State: closed - Opened by danforbes over 1 year ago

#181 - Docs and small refactor for ModelParams

Pull Request - State: closed - Opened by danforbes over 1 year ago

#180 - Add support for newer GGML file types

Pull Request - State: closed - Opened by danforbes over 1 year ago - 1 comment

#179 - README Changes

Pull Request - State: closed - Opened by danforbes over 1 year ago

#178 - Consider current directory if parent directory is blank

Pull Request - State: closed - Opened by tanmaysachan over 1 year ago - 2 comments

#177 - Speed up model loading

Pull Request - State: closed - Opened by tehmatt over 1 year ago - 1 comment

#176 - Documentation additions

Pull Request - State: closed - Opened by danforbes over 1 year ago - 3 comments

#175 - Decrease repetition in evaluation code

Pull Request - State: closed - Opened by danforbes over 1 year ago

#174 - Make it prettier to perform basic inference

Pull Request - State: closed - Opened by danforbes over 1 year ago - 2 comments

#173 - Performance of inference with 65B model on high-end CPU?

Issue - State: open - Opened by vedantroy over 1 year ago - 2 comments

#172 - Load tensors from info rather than manually

Pull Request - State: closed - Opened by danforbes over 1 year ago

#171 - Re-record example GIFs in the README

Issue - State: closed - Opened by philpax over 1 year ago - 1 comment
Labels: documentation, cli

#170 - Added Send + Sync to Model

Pull Request - State: closed - Opened by LLukas22 over 1 year ago

#169 - Cargo install fails "because it has no binaries"

Issue - State: closed - Opened by divinity76 over 1 year ago - 1 comment

#168 - GPT-J model implementation

Pull Request - State: closed - Opened by danforbes over 1 year ago - 2 comments

#167 - GPT-2 segfaults when used through the CLI

Issue - State: closed - Opened by philpax over 1 year ago - 8 comments
Labels: bug, cli

#166 - StableLM Model Support

Pull Request - State: closed - Opened by danforbes over 1 year ago - 3 comments

#165 - Implement Codegen Model

Pull Request - State: closed - Opened by hhamud over 1 year ago - 1 comment

#164 - Fix a bug after bulk rename in README.md

Pull Request - State: closed - Opened by anstadnik over 1 year ago - 2 comments

#163 - BLOOM does not seem to produce meaningful output

Issue - State: closed - Opened by philpax over 1 year ago - 5 comments
Labels: bug, bloom

#162 - Structural Overhaul

Pull Request - State: closed - Opened by danforbes over 1 year ago - 8 comments

#161 - Use a spinner for model loading information

Pull Request - State: closed - Opened by tehmatt over 1 year ago - 2 comments

#160 - mmap with MAP_POPULATE

Pull Request - State: closed - Opened by iacore over 1 year ago - 2 comments

#159 - fix: Proper error message when the model file doesn't exist

Pull Request - State: closed - Opened by tehmatt over 1 year ago - 1 comment

#158 - Add discovered number to magic mismatch error

Pull Request - State: closed - Opened by tehmatt over 1 year ago

#157 - Add multiline support to chat

Pull Request - State: closed - Opened by dustletter over 1 year ago - 2 comments

#156 - fix #155 - multi-line text entry

Pull Request - State: closed - Opened by suzuke over 1 year ago - 3 comments

#155 - Feature request: multi-line text entry

Issue - State: closed - Opened by dustletter over 1 year ago - 2 comments

#154 - Automatic testing with small open-source models

Issue - State: open - Opened by philpax over 1 year ago - 1 comment
Labels: enhancement, maintenance

#153 - Multiple functions accessing the same model

Issue - State: closed - Opened by okpatil4u over 1 year ago - 1 comment

#152 - fix #149 - load tensors by type, ignoring filetype

Pull Request - State: closed - Opened by philpax over 1 year ago

#151 - Related to stop prompt

Issue - State: closed - Opened by hlhr202 over 1 year ago - 5 comments

#150 - A tool to convert multipart models to singlepart models

Issue - State: closed - Opened by philpax over 1 year ago - 4 comments
Labels: enhancement

#149 - GPT4-X cannot be loaded

Issue - State: closed - Opened by philpax over 1 year ago
Labels: bug, model-support

#148 - Add color_eyre for cli error handling

Pull Request - State: closed - Opened by tehmatt over 1 year ago - 2 comments

#147 - Support for SalesForce CodeGen

Issue - State: closed - Opened by philpax over 1 year ago - 5 comments
Labels: enhancement, help wanted, model-support

#146 - Support for StableLM

Issue - State: closed - Opened by philpax over 1 year ago - 1 comment
Labels: enhancement, model-support

#145 - Support for GPT-NeoX

Issue - State: closed - Opened by philpax over 1 year ago - 3 comments
Labels: enhancement, model-support

#144 - Support for GPT-J

Issue - State: closed - Opened by philpax over 1 year ago - 2 comments
Labels: enhancement, model-support

#142 - Support for Dolly models

Issue - State: closed - Opened by vv9k over 1 year ago - 3 comments
Labels: enhancement, model-support

#141 - BLOOM Refactor

Pull Request - State: closed - Opened by danforbes over 1 year ago - 31 comments

#139 - Sync to GGML version as of 20230416 08:55 AM UTC.

Pull Request - State: closed - Opened by KerfuffleV2 over 1 year ago - 1 comment

#138 - Sync to GGML version as of 20230414 06:00 AM UTC.

Pull Request - State: closed - Opened by KerfuffleV2 over 1 year ago - 2 comments

#137 - Build and execute our own computation graph

Issue - State: open - Opened by philpax over 1 year ago - 4 comments
Labels: enhancement, maintenance

#136 - Rename to `llm`

Issue - State: closed - Opened by philpax over 1 year ago - 7 comments
Labels: enhancement, maintenance

#135 - Split up modules

Pull Request - State: closed - Opened by philpax over 1 year ago - 1 comment
Labels: maintenance

#132 - perf: add benchmarks

Issue - State: open - Opened by jon-chuang over 1 year ago - 1 comment
Labels: issue:documentation

#132 - perf: add benchmarks

Issue - State: open - Opened by jon-chuang over 1 year ago - 1 comment
Labels: issue:documentation

#132 - perf: add benchmarks

Issue - State: open - Opened by jon-chuang over 1 year ago - 1 comment
Labels: issue:documentation

#132 - perf: add benchmarks

Issue - State: open - Opened by jon-chuang over 1 year ago - 1 comment
Labels: issue:documentation

#132 - perf: add benchmarks

Issue - State: open - Opened by jon-chuang over 1 year ago - 1 comment
Labels: issue:documentation

#131 - perf: CPU utilization is not 100%

Issue - State: closed - Opened by jon-chuang over 1 year ago - 10 comments

#129 - feat: mmapped ggjt loader

Pull Request - State: closed - Opened by jon-chuang over 1 year ago - 9 comments

#128 - Change Rust-side GGML view ops to take element rather than byte arguments.

Pull Request - State: closed - Opened by KerfuffleV2 over 1 year ago - 4 comments

#125 - Standalone loader

Pull Request - State: closed - Opened by iacore over 1 year ago - 57 comments

#115 - Runs out of ggml context's memory pool on larger models

Issue - State: closed - Opened by juho-p over 1 year ago
Labels: bug

#108 - scripts: use GGMF version of convert-pth-to-ggml.py

Pull Request - State: closed - Opened by philpax over 1 year ago - 1 comment

#107 - Update to latest ggml and bring back bindgen as a dev dependency

Pull Request - State: closed - Opened by setzer22 over 1 year ago - 1 comment

#104 - Fix num_threads flag being ignored

Pull Request - State: closed - Opened by setzer22 over 1 year ago

#100 - Add ggml conversion scripts and documentation for how to convert weights

Pull Request - State: closed - Opened by setzer22 over 1 year ago - 1 comment

#95 - Investigate concurrent inference across threads with one model

Issue - State: closed - Opened by philpax over 1 year ago - 13 comments
Labels: enhancement

#93 - Support the new `mmap`-able ggml format

Issue - State: closed - Opened by philpax over 1 year ago - 12 comments
Labels: enhancement

#92 - Relicense as MIT/Apache2

Issue - State: closed - Opened by philpax over 1 year ago - 11 comments
Labels: maintenance

#90 - Compute perplexity

Issue - State: closed - Opened by philpax over 1 year ago - 2 comments
Labels: enhancement

#89 - Add an example of the ReAct pattern

Issue - State: open - Opened by philpax over 1 year ago - 4 comments
Labels: issue:enhancement, meta:good-first-issue

#86 - Did we support gpt4all-lora-quantized.bin?

Issue - State: closed - Opened by katopz over 1 year ago - 19 comments
Labels: enhancement, model-support

#85 - WIP: Bloom Inference

Pull Request - State: closed - Opened by hhamud over 1 year ago - 10 comments

#84 - Ported `quantize.cpp`

Pull Request - State: closed - Opened by FloppyDisck over 1 year ago - 14 comments

#79 - Parallel loading of the model tensors

Issue - State: open - Opened by philpax over 1 year ago - 5 comments
Labels: enhancement

#77 - Swap strategy for infinite output

Issue - State: open - Opened by setzer22 over 1 year ago - 2 comments
Labels: issue:enhancement

#75 - Support for RWKV

Issue - State: open - Opened by philpax over 1 year ago - 47 comments
Labels: enhancement, model-support

#72 - Embedding extraction

Pull Request - State: closed - Opened by setzer22 over 1 year ago - 16 comments

#57 - Publish to crates.io

Issue - State: closed - Opened by philpax over 1 year ago - 6 comments
Labels: enhancement

#53 - Better multiline support in interactive mode

Issue - State: closed - Opened by 0xAlcibiades over 1 year ago - 3 comments
Labels: enhancement

#45 - Support for BLOOM

Issue - State: closed - Opened by philpax over 1 year ago - 4 comments
Labels: enhancement, model-support

#40 - Convert `quantize.cpp` to Rust

Issue - State: closed - Opened by philpax over 1 year ago - 5 comments
Labels: enhancement

#37 - feat: HTTP server for streaming inferences

Pull Request - State: closed - Opened by jack-michaud over 1 year ago - 12 comments

#36 - Adds a way to specify the seed for RNG

Pull Request - State: closed - Opened by setzer22 over 1 year ago

#32 - Update GGML

Pull Request - State: closed - Opened by setzer22 over 1 year ago

#31 - Non-`ggml` backend

Issue - State: open - Opened by philpax over 1 year ago - 28 comments
Labels: enhancement, help wanted

#20 - Automatic builds of `llama[-rs]-cli` binaries to download

Issue - State: closed - Opened by philpax over 1 year ago - 1 comment
Labels: enhancement

#16 - Real-time chat platform

Issue - State: closed - Opened by philpax over 1 year ago - 6 comments

#14 - Implementation of prompt caching

Pull Request - State: closed - Opened by setzer22 over 1 year ago - 4 comments

GitHub / rustformers/llm issues and pull requests