Merge branch 'main' into feat/together-interpreter

Author: CompuIves

.devcontainer/Dockerfile

@@ -1,4 +1,4 @@
-FROM mcr.microsoft.com/devcontainers/python:3.9
+FROM mcr.microsoft.com/devcontainers/python:3.10
 
 ENV PYTHONUNBUFFERED 1
 

.github/workflows/_integration_tests.yml

@@ -3,28 +3,27 @@ name: integration-tests
 on:
   push:
     branches: [ main ]
-    paths:
-      - "**.py"
+    paths-ignore:
+      - "**.md"
 
   pull_request:
     branches: [ main ]
-    paths:
-      - "**.py"
+    paths-ignore:
+      - "**.md"
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
   cancel-in-progress: true
 
 env:
-  POETRY_VERSION: "1.7.1"
+  POETRY_VERSION: "2.1.1"
 
 jobs:
   build:
     runs-on: ubuntu-latest
     strategy:
       matrix:
         python-version:
-          - "3.9"
           - "3.10"
           - "3.11"
           - "3.12"
@@ -39,14 +38,10 @@ jobs:
           poetry-version: ${{ env.POETRY_VERSION }}
           cache-key: core
 
-      - name: Install dependencies
+      - name: Install together with dependencies
         shell: bash
         run: poetry install --with quality,tests
 
-      - name: Install together
-        run: |
-          poetry run pip install .
-
       - name: Run integration tests
         shell: bash
         env:

.github/workflows/_tests.yml

@@ -3,13 +3,13 @@ name: unit-tests
 on:
   push:
     branches: [ main ]
-    paths:
-      - "**.py"
+    paths-ignore:
+      - "**.md"
 
   pull_request:
     branches: [ main ]
-    paths:
-      - "**.py"
+    paths-ignore:
+      - "**.md"
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
@@ -24,7 +24,6 @@ jobs:
     strategy:
       matrix:
         python-version:
-          - "3.9"
           - "3.10"
           - "3.11"
           - "3.12"

.github/workflows/check_code_quality.yml

@@ -3,13 +3,9 @@ name: pre-commit
 on:
   push:
     branches: [ main ]
-    paths:
-      - "**.py"
 
   pull_request:
     branches: [ main ]
-    paths:
-      - "**.py"
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}

.github/workflows/upload-to-pypi.yml

@@ -4,6 +4,9 @@ on:
   release:
     types: [published]
 
+env:
+  POETRY_VERSION: "2.1.1"
+
 jobs:
   build:
     runs-on: ubuntu-latest
@@ -13,16 +16,15 @@ jobs:
 
     steps:
     - uses: actions/checkout@v4
-    - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v2.2.2
+    - name: Set up Python ${{ matrix.python-version }} + Poetry ${{ env.POETRY_VERSION }}
+      uses: "./.github/actions/poetry_setup"
       with:
         python-version: ${{ matrix.python-version }}
-
-    - name: Install poetry
-      run: curl -sSL https://install.python-poetry.org | python3 -
+        poetry-version: ${{ env.POETRY_VERSION }}
+        cache-key: core
 
     - name: Set up cache
-      uses: actions/cache@v2.1.6
+      uses: actions/cache@v4
       with:
         path: .venv
         key: venv-${{ matrix.python-version }}-${{ hashFiles('pyproject.toml') }}-${{ hashFiles('poetry.lock') }}

README.md

@@ -10,7 +10,7 @@
 [![Discord](https://dcbadge.vercel.app/api/server/9Rk6sSeWEG?style=flat&compact=true)](https://discord.com/invite/9Rk6sSeWEG)
 [![Twitter](https://img.shields.io/twitter/url/https/twitter.com/togethercompute.svg?style=social&label=Follow%20%40togethercompute)](https://twitter.com/togethercompute)
 
-The [Together Python API Library](https://pypi.org/project/together/) is the official Python client for Together's API platform, providing a convenient way for interacting with the REST APIs and enables easy integrations with Python 3.8+ applications with easy to use synchronous and asynchronous clients.
+The [Together Python API Library](https://pypi.org/project/together/) is the official Python client for Together's API platform, providing a convenient way for interacting with the REST APIs and enables easy integrations with Python 3.10+ applications with easy to use synchronous and asynchronous clients.
 
 
 
@@ -67,7 +67,7 @@ print(response.choices[0].message.content)
 response = client.chat.completions.create(
     model="meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo",
     messages=[{
-        "role": "user", 
+        "role": "user",
         "content": [
             {
                 "type": "text",
@@ -91,7 +91,7 @@ response = client.chat.completions.create(
         "role": "user",
         "content": [
             {
-                "type": "text", 
+                "type": "text",
                 "text": "Compare these two images."
             },
             {

poetry.lock

@@ -12,7 +12,7 @@ build-backend = "poetry.masonry.api"
 
 [tool.poetry]
 name = "together"
-version = "1.4.2"
+version = "1.5.3"
 authors = [
     "Together AI <support@together.ai>"
 ]
@@ -28,7 +28,7 @@ repository = "https://github.com/togethercomputer/together-python"
 homepage = "https://github.com/togethercomputer/together-python"
 
 [tool.poetry.dependencies]
-python = "^3.9"
+python = "^3.10"
 typer = ">=0.9,<0.16"
 requests = "^2.31.0"
 rich = "^13.8.1"

pyproject.toml

@@ -127,6 +127,11 @@ def endpoints(ctx: click.Context) -> None:
     is_flag=True,
     help="Create the endpoint in STOPPED state instead of auto-starting it",
 )
+@click.option(
+    "--inactive-timeout",
+    type=int,
+    help="Number of minutes of inactivity after which the endpoint will be automatically stopped. Set to 0 to disable.",
+)
 @click.option(
     "--wait",
     is_flag=True,
@@ -146,6 +151,7 @@ def create(
     no_prompt_cache: bool,
     no_speculative_decoding: bool,
     no_auto_start: bool,
+    inactive_timeout: int | None,
     wait: bool,
 ) -> None:
     """Create a new dedicated inference endpoint."""
@@ -170,6 +176,7 @@ def create(
             disable_prompt_cache=no_prompt_cache,
             disable_speculative_decoding=no_speculative_decoding,
             state="STOPPED" if no_auto_start else "STARTED",
+            inactive_timeout=inactive_timeout,
         )
     except InvalidRequestError as e:
         print_api_error(e)
@@ -194,6 +201,8 @@ def create(
         click.echo("  Speculative decoding: disabled", err=True)
     if no_auto_start:
         click.echo("  Auto-start: disabled", err=True)
+    if inactive_timeout is not None:
+        click.echo(f"  Inactive timeout: {inactive_timeout} minutes", err=True)
 
     click.echo(f"Endpoint created successfully, id: {response.id}", err=True)
 
@@ -371,6 +380,11 @@ def list(
     type=int,
     help="New maximum number of replicas to scale up to",
 )
+@click.option(
+    "--inactive-timeout",
+    type=int,
+    help="Number of minutes of inactivity after which the endpoint will be automatically stopped. Set to 0 to disable.",
+)
 @click.pass_obj
 @handle_api_errors
 def update(
@@ -379,9 +393,10 @@ def update(
     display_name: str | None,
     min_replicas: int | None,
     max_replicas: int | None,
+    inactive_timeout: int | None,
 ) -> None:
     """Update a dedicated inference endpoint's configuration."""
-    if not any([display_name, min_replicas, max_replicas]):
+    if not any([display_name, min_replicas, max_replicas, inactive_timeout]):
         click.echo("Error: At least one update option must be specified", err=True)
         sys.exit(1)
 
@@ -400,6 +415,8 @@ def update(
     if min_replicas is not None and max_replicas is not None:
         kwargs["min_replicas"] = min_replicas
         kwargs["max_replicas"] = max_replicas
+    if inactive_timeout is not None:
+        kwargs["inactive_timeout"] = inactive_timeout
 
     _response = client.endpoints.update(endpoint_id, **kwargs)
 
@@ -410,6 +427,8 @@ def update(
     if min_replicas is not None and max_replicas is not None:
         click.echo(f"  Min replicas: {min_replicas}", err=True)
         click.echo(f"  Max replicas: {max_replicas}", err=True)
+    if inactive_timeout is not None:
+        click.echo(f"  Inactive timeout: {inactive_timeout} minutes", err=True)
 
     click.echo("Successfully updated endpoint", err=True)
     click.echo(endpoint_id)