Prepare v2.13.27 release.

Update AbstractVision/Voice/Music capability plugin floors to the latest released versions and refresh packaging/docs references.

Co-authored-by: Cursor <cursoragent@cursor.com>

Author: lpalbou

CHANGELOG.md

@@ -7,6 +7,14 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [2.13.27] - 2026-05-23
+
+### Changed
+- **Capability plugin floors**: updated optional capability plugin install floors to `abstractvoice>=0.10.15`, `abstractvision>=0.3.13`, and `abstractmusic>=0.1.12` (plus matching turnkey profiles and docs references).
+
+### Verified
+- **Hermetic test suite**: `pytest` passes with local/provider/live tests disabled (CI-style defaults).
+
 ## [2.13.26] - 2026-05-23
 
 ### Changed

README.md

@@ -358,7 +358,7 @@ Text-only `generate(...)` is unchanged. For advanced/provider-specific work,
 the direct `llm.vision.*`, `llm.voice.*`, `llm.audio.*`, and `llm.music.*` facades remain
 available. Configure `abstractvision` and `abstractvoice` backends first for
 real generation; configure `abstractmusic` for music generation. With
-`abstractmusic>=0.1.8`, the default music backend is the lightweight remote
+`abstractmusic>=0.1.12`, the default music backend is the lightweight remote
 ACE Music path; set `ACEMUSIC_API_KEY` before use. Local music engines remain
 optional plugin extras.
 

abstractcore/server/README.md

@@ -62,7 +62,7 @@ Runtime note: omitted `task` on `/acore/models/load` keeps the existing text-gen
 | **Audio Translations** | `/v1/audio/translations` | `file` | Translate audio (not yet supported) |
 | **Text-to-Speech** | `/v1/audio/speech`, `/{provider}/v1/audio/speech` | `input`, optional `model`/`provider`/`base_url` | Generate audio (optional) |
 | **Voice Clone** | `/v1/voice/clone`, `/{provider}/v1/voice/clone` | `file`, optional `model`/`provider`/`base_url` | Create a compatible custom voice (optional) |
-| **Text-to-Music** | `/v1/audio/music` | `prompt` | Generate music/audio through optional AbstractMusic backends; `abstractmusic>=0.1.8` supports remote ACE Music with `ACEMUSIC_API_KEY` |
+| **Text-to-Music** | `/v1/audio/music` | `prompt` | Generate music/audio through optional AbstractMusic backends; `abstractmusic>=0.1.12` supports remote ACE Music with `ACEMUSIC_API_KEY` |
 | **Documents** | `/v1/chat/completions` | `content: [text, file_url]` | PDF/CSV processing |
 | **Tools** | `/v1/chat/completions` | `tools`, `tool_choice` | Function calling |
 | **Embeddings** | `/v1/embeddings` | `model`, `input` | Text embeddings |

abstractcore/utils/version.py

@@ -10,4 +10,4 @@
 """
 
 # Package version - update this when releasing new versions
-__version__ = "2.13.26"
+__version__ = "2.13.27"

docs/capabilities.md

@@ -123,7 +123,7 @@ The server exposes the same deep catalogs through:
 - `GET /v1/audio/music/providers`
 - `GET /v1/audio/music/models`
 
-For `abstractmusic>=0.1.8`, the default lightweight music backend is the
+For `abstractmusic>=0.1.12`, the default lightweight music backend is the
 remote ACE Music API path (`provider="acemusic"` or `/acemusic/v1/audio/music`).
 Set `ACEMUSIC_API_KEY` in the server or Python environment. The server music
 route accepts `wav`, `mp3`, and `flac`; individual backends may support fewer

docs/huggingface-model-compatibility.md

@@ -21,7 +21,7 @@ install every optional Transformers quantization runtime, because those runtimes
 platform-specific and can carry dependency pins that conflict with the rest of the local stack.
 Fresh installs resolve the newest compatible Transformers release allowed by AbstractCore's
 dependency range. Very new architectures such as Gemma4 require a recent Transformers build.
-Audio/voice extras use `abstractvoice>=0.10.14` and `omnivoice>=0.1.5`.
+Audio/voice extras use `abstractvoice>=0.10.15` and `omnivoice>=0.1.5`.
 
 ## Quantized Transformers Checkpoints
 

docs/server.md

@@ -743,7 +743,7 @@ The returned `voice_id` / `id` can be used as the `voice` value in
 | `response_format` or `format` | no | Server contract supports `wav`, `mp3`, and `flac`; backend support can be narrower. |
 | extra top-level fields | no | Best-effort passthrough to the installed music capability plugin. |
 
-With `abstractmusic>=0.1.8`, the base install includes the remote ACE Music
+With `abstractmusic>=0.1.12`, the base install includes the remote ACE Music
 backend. Configure `ACEMUSIC_API_KEY` in the server environment, optionally set
 `ACEMUSIC_BASE_URL`, and use `provider="acemusic"` or the `/acemusic/v1/audio/music`
 path. Local ACE-Step/Diffusers routes remain opt-in AbstractMusic extras.

llms-full.txt

@@ -3,7 +3,7 @@
 > This is the "full context" companion to `llms.txt`: a single, self-contained guide for agents and developers using AbstractCore. It intentionally avoids GitHub link hubs; only a few external references are included (API-key signup pages).
 
 Last updated: 2026-05-23
-Package version: 2.13.26
+Package version: 2.13.27
 Supported Python: 3.9+; GitHub CI tests 3.9, 3.10, 3.11, 3.12, and 3.13.
 
 ## How to use this file (agents)
@@ -651,7 +651,7 @@ AbstractCore supports optional capability plugins discovered via Python entry po
 - install `abstractvoice` -> enables `llm.voice` and `llm.audio` (TTS/STT)
 - install `abstractvision` -> enables `llm.vision` (generative vision; often via an OpenAI-compatible images endpoint)
 - install `abstractcore[music]` -> enables `llm.music` and music output routing for text-to-music through `abstractmusic`
-- current plugin floors in this source: `abstractvoice>=0.10.14`, `abstractvision>=0.3.9`, `abstractmusic>=0.1.8`
+- current plugin floors in this source: `abstractvoice>=0.10.15`, `abstractvision>=0.3.13`, `abstractmusic>=0.1.12`
 
 Server note:
 - the server can optionally expose `/v1/images/*` (requires `abstractvision`)
@@ -10202,7 +10202,7 @@ python -m abstractcore.utils.cli --prompt "What's in @document.pdf and @image.jp
 To keep the default `abstractcore` install dependency-light while still enabling deterministic modality APIs, AbstractCore supports optional **capability plugins**:
 - `abstractvoice` provides `core.voice` + `core.audio` (TTS/STT).
 - `abstractvision` provides `core.vision` (T2I/I2I/T2V/I2V; backend-pluggable).
-- `abstractmusic` can provide `core.music` (text-to-music) when installed; `abstractmusic>=0.1.8` includes the remote ACE Music backend through `ACEMUSIC_API_KEY`.
+- `abstractmusic` can provide `core.music` (text-to-music) when installed; `abstractmusic>=0.1.12` includes the remote ACE Music backend through `ACEMUSIC_API_KEY`.
 
 Discovery:
 - `llm.capabilities.status()` returns a JSON-safe snapshot (which backends are available/selected, plus install hints).

llms.txt

@@ -2,7 +2,7 @@
 
 > Unified Python interface for cloud + local LLM providers with streaming, tool calling, structured output, media handling (images/audio/video + documents), embeddings, and an optional OpenAI-compatible HTTP server. Default install stays lightweight; features are enabled via extras.
 
-Last reviewed against source: 2026-05-23. Package version in source: 2.13.26. Supported Python: 3.9+; GitHub CI tests 3.9, 3.10, 3.11, 3.12, and 3.13.
+Last reviewed against source: 2026-05-23. Package version in source: 2.13.27. Supported Python: 3.9+; GitHub CI tests 3.9, 3.10, 3.11, 3.12, and 3.13.
 
 This is the curated, high-signal index for agents. For a self-contained handbook with the important details inlined, read `llms-full.txt`.
 
@@ -61,7 +61,7 @@ Quick start:
 - [Provider registry](abstractcore/providers/registry.py): provider IDs, defaults, install extras
 - [Provider setup guide](docs/prerequisites.md): env vars + examples for OpenAI/Anthropic/Ollama/LMStudio/MLX/HuggingFace/vLLM/OpenRouter/Portkey/OpenAI-compatible
 - Install profile vocabulary: `abstractcore[apple]` is the hardware alias for the MLX local LLM stack, `abstractcore[gpu]` is the hardware alias for the vLLM local LLM stack, and `abstractcore[all-apple]` / `abstractcore[all-gpu]` are larger aggregate profiles.
-- Optional media plugin floors in this source: `abstractvoice>=0.10.14`, `abstractvision>=0.3.9`, and `abstractmusic>=0.1.8`.
+- Optional media plugin floors in this source: `abstractvoice>=0.10.15`, `abstractvision>=0.3.13`, and `abstractmusic>=0.1.12`.
 
 ## Server (OpenAI-compatible `/v1`)
 

pyproject.toml

@@ -214,12 +214,12 @@ media = [
 ]
 
 voice = [
-    "abstractvoice>=0.10.14",
+    "abstractvoice>=0.10.15",
     "omnivoice>=0.1.5",
 ]
 
 audio = [
-    "abstractvoice>=0.10.14",
+    "abstractvoice>=0.10.15",
     "omnivoice>=0.1.5",
 ]
 
@@ -266,15 +266,15 @@ all = [
     "pymupdf-layout>=1.26.6,<2.0.0",
     "unstructured[docx,pptx,xlsx,odt,rtf]>=0.10.0,<1.0.0",
     "pandas>=1.0.0,<3.0.0",
-    "abstractvoice>=0.10.14",
+    "abstractvoice>=0.10.15",
     "omnivoice>=0.1.5",
-    "abstractmusic>=0.1.8",
+    "abstractmusic>=0.1.12",
     # Server
     "fastapi>=0.100.0,<1.0.0",
     "uvicorn[standard]>=0.23.0,<1.0.0",
     "python-multipart>=0.0.6,<1.0.0",
     "sse-starlette>=1.6.0,<2.0.0",
-    "abstractvision>=0.3.9",
+    "abstractvision>=0.3.13",
 ]
 
 all-apple = [
@@ -309,15 +309,15 @@ all-apple = [
     "pymupdf-layout>=1.26.6,<2.0.0",
     "unstructured[docx,pptx,xlsx,odt,rtf]>=0.10.0,<1.0.0",
     "pandas>=1.0.0,<3.0.0",
-    "abstractvoice[all-apple]>=0.10.14",
+    "abstractvoice[all-apple]>=0.10.15",
     "omnivoice>=0.1.5",
-    "abstractmusic[all-apple]>=0.1.8",
+    "abstractmusic[all-apple]>=0.1.12",
     # Server
     "fastapi>=0.100.0,<1.0.0",
     "uvicorn[standard]>=0.23.0,<1.0.0",
     "python-multipart>=0.0.6,<1.0.0",
     "sse-starlette>=1.6.0,<2.0.0",
-    "abstractvision[all-apple]>=0.3.9",
+    "abstractvision[all-apple]>=0.3.13",
 ]
 
 all-gpu = [
@@ -348,15 +348,15 @@ all-gpu = [
     "pymupdf-layout>=1.26.6,<2.0.0",
     "unstructured[docx,pptx,xlsx,odt,rtf]>=0.10.0,<1.0.0",
     "pandas>=1.0.0,<3.0.0",
-    "abstractvoice[all-gpu]>=0.10.14",
+    "abstractvoice[all-gpu]>=0.10.15",
     "omnivoice>=0.1.5",
-    "abstractmusic[all-gpu]>=0.1.8",
+    "abstractmusic[all-gpu]>=0.1.12",
     # Server
     "fastapi>=0.100.0,<1.0.0",
     "uvicorn[standard]>=0.23.0,<1.0.0",
     "python-multipart>=0.0.6,<1.0.0",
     "sse-starlette>=1.6.0,<2.0.0",
-    "abstractvision[all-gpu]>=0.3.9",
+    "abstractvision[all-gpu]>=0.3.13",
 ]
 
 # Legacy broad bundle kept for existing users. New installs should usually
@@ -388,15 +388,15 @@ all-non-mlx = [
     "pymupdf-layout>=1.26.6,<2.0.0",
     "unstructured[docx,pptx,xlsx,odt,rtf]>=0.10.0,<1.0.0",
     "pandas>=1.0.0,<3.0.0",
-    "abstractvoice>=0.10.14",
+    "abstractvoice>=0.10.15",
     "omnivoice>=0.1.5",
-    "abstractmusic>=0.1.8",
+    "abstractmusic>=0.1.12",
     # Server
     "fastapi>=0.100.0,<1.0.0",
     "uvicorn[standard]>=0.23.0,<1.0.0",
     "python-multipart>=0.0.6,<1.0.0",
     "sse-starlette>=1.6.0,<2.0.0",
-    "abstractvision>=0.3.9",
+    "abstractvision>=0.3.13",
 ]
 
 # Development dependencies
@@ -425,23 +425,23 @@ server = [
 # Vision (optional) - delegates to AbstractVision.
 # Kept as a separate extra for installs that want vision without the HTTP server.
 vision = [
-    "abstractvision>=0.3.9",
+    "abstractvision>=0.3.13",
 ]
 
 vision-diffusers = [
-    "abstractvision[huggingface]>=0.3.9",
+    "abstractvision[huggingface]>=0.3.13",
 ]
 
 vision-sdcpp = [
-    "abstractvision[sdcpp]>=0.3.9",
+    "abstractvision[sdcpp]>=0.3.13",
 ]
 
 vision-local = [
-    "abstractvision[local]>=0.3.9",
+    "abstractvision[local]>=0.3.13",
 ]
 
 music = [
-    "abstractmusic>=0.1.8",
+    "abstractmusic>=0.1.12",
 ]
 
 # Testing dependencies
@@ -506,14 +506,14 @@ full-dev = [
     "pymupdf-layout>=1.26.6,<2.0.0",
     "unstructured[docx,pptx,xlsx,odt,rtf]>=0.10.0,<1.0.0",
     "pandas>=1.0.0,<3.0.0",
-    "abstractvoice>=0.10.14",
+    "abstractvoice>=0.10.15",
     "omnivoice>=0.1.5",
-    "abstractmusic>=0.1.8",
+    "abstractmusic>=0.1.12",
     "fastapi>=0.100.0,<1.0.0",
     "uvicorn[standard]>=0.23.0,<1.0.0",
     "python-multipart>=0.0.6,<1.0.0",
     "sse-starlette>=1.6.0,<2.0.0",
-    "abstractvision>=0.3.9",
+    "abstractvision>=0.3.13",
     # Tooling
     "pytest>=7.0.0",
     "pytest-asyncio>=0.21.0",