This is an automated email from the ASF dual-hosted git repository.
tqchen pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/tvm-ffi.git
The following commit(s) were added to refs/heads/main by this push:
new aeaeca0 Update readme text (#182)
aeaeca0 is described below
commit aeaeca0af7143d905714c99702493a441ca92726
Author: Tianqi Chen <[email protected]>
AuthorDate: Mon Oct 20 22:13:45 2025 -0700
Update readme text (#182)
Co-authored-by: gemini-code-assist[bot]
<176961590+gemini-code-assist[bot]@users.noreply.github.com>
---
.asf.yaml | 2 +-
README.md | 38 +++++++++++++++++---------------------
2 files changed, 18 insertions(+), 22 deletions(-)
diff --git a/.asf.yaml b/.asf.yaml
index 1d179c9..c6b24cb 100644
--- a/.asf.yaml
+++ b/.asf.yaml
@@ -16,7 +16,7 @@
# under the License.
github:
- description: "TVM FFI"
+ description: "Open ABI and FFI for Machine Learning Systems"
homepage: https://tvm.apache.org/ffi
labels:
- ffi
diff --git a/README.md b/README.md
index 48cfefa..f238e7c 100644
--- a/README.md
+++ b/README.md
@@ -22,34 +22,30 @@
Apache TVM FFI is an open ABI and FFI for machine learning systems. It is a
minimal, framework-agnostic,
yet flexible open convention with the following systems in mind:
-- Kernel libraries: ship one wheel to support multiple frameworks, Python
versions, and different languages.
-- Kernel DSLs: reusable open ABI for JIT and AOT kernel exposure to PyTorch,
JAX, and other ML runtimes.
-- ML frameworks and runtimes: unified mechanism to connect libraries and DSLs
that adopt the ABI convention.
-- Coding agents: unified mechanism to package and ship generated code to
production environments.
-- ML infrastructure: cross-language support for Python, C++, and Rust, and
DSLs.
+- **Kernel libraries** - ship one wheel to support multiple frameworks, Python
versions, and different languages.
+- **Kernel DSLs** - reusable open ABI for JIT and AOT kernel exposure
frameworks and runtimes.
+- **Frameworks and runtimes** - a uniform extension point for ABI-compliant
libraries and DSLs.
+- **ML infrastructure** - out-of-the-box bindings and interop for Python, C++,
and Rust.
+- **Coding agents** - a unified mechanism for shipping generated code in
production.
-It has the following technical features:
+## Features
-- DLPack-compatible Tensor data ABI to seamlessly support many frameworks such
as PyTorch, JAX, CuPy and others that support DLPack convention.
-- Compact value and function calling convention for common data types in
machine learning.
-- Stable, minimal, and flexible C ABI to support machine learning system
use-cases.
-- Out-of-the-box multi-language support for Python, C++, Rust, and future path
for other languages.
+* **Stable, minimal C ABI** designed for kernels, DSLs, and runtime
extensibility.
+* **Zero-copy interop** across PyTorch, JAX, and CuPy using [DLPack
protocol](https://data-apis.org/array-api/2024.12/design_topics/data_interchange.html).
+* **Compact value and call convention** covering common data types for ultra
low-overhead ML applications.
+* **Multi-language support** out of the box: Python, C++, and Rust (with a
path towards more languages).
-With these technical solutions, we can enable better **interoperability**
across machine learning frameworks,
-libraries, kernel DSLs, and coding agents, **ship one wheel** to support
multiple frameworks and Python versions (including free-threaded python),
-and build infrastructure solutions across environments.
+These enable broad **interoperability** across frameworks, libraries, DSLs,
and agents; the ability to **ship one wheel** for multiple frameworks and
Python versions (including free-threaded Python); and consistent infrastructure
across environments.
## Status and Release Versioning
-C ABI stability is the top priority of this effort. We also prioritize
minimalism and
-efficiency in the core so it is portable and can be used broadly.
-We are current in the RFC stage, which means the main features are complete
and ABI stable.
-We also recognize potential needs for evolution to ensure it works best for
the machine
-learning systems community, and would like to work together collectively with
the community for such evolution.
-The RFC stage is a period where we are working with the open source communities
-to ensure we evolve the ABI to meet the potential needs of frameworks.
+**C ABI stability** is our top priority.
+
+**Status: RFC** Main features are complete and ABI stable. We recognize
potential needs for evolution to ensure
+it works best for the machine learning systems community, and would like to
work together with the
+community for such evolution. We plan to stay in the RFC stage for three
months from the v0.1.0 release.
Releases during the RFC stage will be `0.X.Y`, where bumps in `X` indicate C
ABI-breaking changes
-and `Y` indicates other changes. We anticipate the RFC stage will last for a
few months, then we will start to follow
+and `Y` indicates other changes. We anticipate the RFC stage will last for
three months, then we will start following
[Semantic
Versioning](https://packaging.python.org/en/latest/discussions/versioning/)
(`major.minor.patch`) going forward.
