These 25 commits are when the Protocol Buffers files have changed:
| Commit: | e45caef | |
|---|---|---|
| Author: | Gato | |
| Committer: | GitHub | |
feat(api,vllm,llama-cpp-python): add token counting feature with gRPC support (#964) * Updates grpc generated code * Adds token counting endpoint (leapfrogai/v1/count/tokens), grpc definitions, and updates backends to allow for token counting
The documentation is generated from this commit.
| Commit: | 014329c | |
|---|---|---|
| Author: | Justin Law | |
| Committer: | GitHub | |
chore(sdk, api): fix types, enhance chat, audio and completion tests (#1038) ### BREAKING CHANGES - fixes `FinishReason` to be an enum in both Completions and ChatCompletions protobufs - modifies API gRPC handler, typing utils, and helper utils to use a `Enum` class to define and transform the stub responses - fixes `object` and `created` field for `Completions` type as defined in the [OpenAI API specification](https://platform.openai.com/docs/api-reference/completions) - uses Literal `text_completion` for `object` - uses `ChatCompletions` type's `created` for `Completions` `created`
| Commit: | 4e5c1b6 | |
|---|---|---|
| Author: | Ed | |
| Committer: | GitHub | |
fix(sdk): best_of in completions was an int and should be a string (#1031) * changing best_of to a string in completions * Revert "changing best_of to a string in completions" This reverts commit 1166dd82970fac7ccdeb183183cc0dbd4db71664. Needed to ensure pre-commit was on firstly * update completion proto and ran make gen-python * update ruff configurations * drop unsupported property * fix formatting
| Commit: | d25bfd7 | |
|---|---|---|
| Author: | John Alling | |
| Committer: | GitHub | |
choer: refactor project structure to handle multiple pyprojects in monorepo (#361) * break out pyproject into each package * minor fixes: package build dests, docs, remove unused deploy values * minor config and docs changes * move sdk out of leapfrogai_api * fully separate api and sdk * replace components packages reference to sdk * Makefile and README clarifications * update e2e workflow * add whisper dev dependencies * update make clean target to remove src and package builds * Update pyproject descriptions and documentation * remove sdk package versions from pyprojects * use pinned dev dependencies in whisper build * simplify docker wheel builds --------- Co-authored-by: Jon Perry <yrrepnoj@gmail.com>
| Commit: | 70e0bdf | |
|---|---|---|
| Author: | Jonathan Perry | |
| Committer: | GitHub | |
chore: Consolidate our repos into a monorepo (#268) * chore: migrate api and sdk-typing code to this repo * chore: add api Dockerfile and Zarf configuration * fix: tinker with versions for api zarf configuration * fix: set ModelConfig defaul for BackendConfig class * chore: update api package README * fix: remove generated python code for gitignore
| Commit: | 60c7a5f | |
|---|---|---|
| Author: | Gregory | |
| Committer: | GitHub | |
rewrote README and deleted remainder of monorepo that has been broken out into separate repos (#247)
This commit does not contain any .proto files.
| Commit: | 4a63bf8 | |
|---|---|---|
| Author: | Gerred Dillon | |
fix proto paths
| Commit: | 50a62b5 | |
|---|---|---|
| Author: | Colin | |
feat: rename chat.proto's CompletionChoice to ChatCompletionChoice avoid naming collisions with completion.proto's CompletionChoice
| Commit: | 1bbabb9 | |
|---|---|---|
| Author: | Colin | |
feat: factor chat proto into separate services for steam and non-stream regen go and python code + add chat code generation to the Makefile
| Commit: | 0e7a46c | |
|---|---|---|
| Author: | Colin | |
feat: updated chat completion proto
| Commit: | b320868 | |
|---|---|---|
| Author: | Colin | |
feat: renamed generate endpoint to completion
| Commit: | f537f26 | |
|---|---|---|
| Author: | Colin | |
feat: add chat proto and mpt model impl
| Commit: | 5e8dd4b | |
|---|---|---|
| Author: | Colin | |
feat: add chat and completion pbs
| Commit: | 1f94af7 | |
|---|---|---|
| Author: | Gerred Dillon | |
wip: ctransformers
| Commit: | 626152f | |
|---|---|---|
| Author: | Tom Runyon | |
api fixes Signed-off-by: Tom Runyon <tom@defenseunicorns.com>
| Commit: | f899fed | |
|---|---|---|
| Author: | Gerred Dillon | |
Make whisper work and update protobufs
| Commit: | 62fd6a3 | |
|---|---|---|
| Author: | Tom Runyon | |
embedding working Signed-off-by: Tom Runyon <tom@defenseunicorns.com>
| Commit: | 4c22033 | |
|---|---|---|
| Author: | Tom Runyon | |
Audio work Signed-off-by: Tom Runyon <tom@defenseunicorns.com>
| Commit: | 5ebae85 | |
|---|---|---|
| Author: | Tom Runyon | |
first skeleton at go-api server Signed-off-by: Tom Runyon <tom@defenseunicorns.com>
| Commit: | 21652d3 | |
|---|---|---|
| Author: | Gerred Dillon | |
Move models into a single subfolder, add minilm2
| Commit: | 14315ee | |
|---|---|---|
| Author: | Gerred Dillon | |
incremental commit, change API server itself to use leapfrog
| Commit: | 9cf9383 | |
|---|---|---|
| Author: | Tom Runyon | |
trying a new module layout Signed-off-by: Tom Runyon <tom@defenseunicorns.com>
| Commit: | 502bc57 | |
|---|---|---|
| Author: | Tom Runyon | |
split out protos Signed-off-by: Tom Runyon <tom@defenseunicorns.com>
| Commit: | 05b148d | |
|---|---|---|
| Author: | Tom Runyon | |
first proto example Signed-off-by: Tom Runyon <tom@defenseunicorns.com>
| Commit: | fa8d800 | |
|---|---|---|
| Author: | Tom Runyon | |
initial checkin for deployment Signed-off-by: Tom Runyon <tom@defenseunicorns.com>