<feed xmlns='http://www.w3.org/2005/Atom'>
<title>ffwd/pplx_embed_mlx.h, branch master</title>
<subtitle>Fastforward embedding inference engine</subtitle>
<id>https://git.onurgunduz.com/ffwd/atom/pplx_embed_mlx.h?h=master</id>
<link rel='self' href='https://git.onurgunduz.com/ffwd/atom/pplx_embed_mlx.h?h=master'/>
<link rel='alternate' type='text/html' href='https://git.onurgunduz.com/ffwd/'/>
<updated>2026-06-14T06:27:03Z</updated>
<entry>
<title>gitignore: track new binary and library names</title>
<updated>2026-06-14T06:27:03Z</updated>
<author>
<name>Onur Gündüz</name>
<email>ogunduz@gmail.com</email>
</author>
<published>2026-06-10T06:24:39Z</published>
<link rel='alternate' type='text/html' href='https://git.onurgunduz.com/ffwd/commit/?id=2c37727e653d2340d8977dd5db4addc767dd81ea'/>
<id>urn:sha1:2c37727e653d2340d8977dd5db4addc767dd81ea</id>
<content type='text'>
</content>
</entry>
<entry>
<title>mlx: keep late maxsim device resident</title>
<updated>2026-06-14T06:27:03Z</updated>
<author>
<name>Onur Gündüz</name>
<email>ogunduz@gmail.com</email>
</author>
<published>2026-06-03T17:18:09Z</published>
<link rel='alternate' type='text/html' href='https://git.onurgunduz.com/ffwd/commit/?id=e751321fc00e07aeeb9337f5f21b7191b051c08a'/>
<id>urn:sha1:e751321fc00e07aeeb9337f5f21b7191b051c08a</id>
<content type='text'>
</content>
</entry>
<entry>
<title>late: add mlx token vector support</title>
<updated>2026-06-14T06:27:03Z</updated>
<author>
<name>Onur Gündüz</name>
<email>ogunduz@gmail.com</email>
</author>
<published>2026-06-03T09:45:40Z</published>
<link rel='alternate' type='text/html' href='https://git.onurgunduz.com/ffwd/commit/?id=7cc87daedf764da43c338ba1cba739bd6af6e950'/>
<id>urn:sha1:7cc87daedf764da43c338ba1cba739bd6af6e950</id>
<content type='text'>
</content>
</entry>
<entry>
<title>mlx: add opt-in 8-bit weight quantization</title>
<updated>2026-06-14T06:27:03Z</updated>
<author>
<name>Onur Gündüz</name>
<email>ogunduz@gmail.com</email>
</author>
<published>2026-06-02T22:30:12Z</published>
<link rel='alternate' type='text/html' href='https://git.onurgunduz.com/ffwd/commit/?id=9d69c13793c4c69c080ed4273bcd36ded69de2bb'/>
<id>urn:sha1:9d69c13793c4c69c080ed4273bcd36ded69de2bb</id>
<content type='text'>
</content>
</entry>
<entry>
<title>server: batch contextual document inference</title>
<updated>2026-06-14T06:27:03Z</updated>
<author>
<name>Onur Gündüz</name>
<email>ogunduz@gmail.com</email>
</author>
<published>2026-05-30T22:39:10Z</published>
<link rel='alternate' type='text/html' href='https://git.onurgunduz.com/ffwd/commit/?id=27e76c88f0c1b33cba54dd31fade9b044719c8bc'/>
<id>urn:sha1:27e76c88f0c1b33cba54dd31fade9b044719c8bc</id>
<content type='text'>
</content>
</entry>
<entry>
<title>dist: load distributed layer shards selectively</title>
<updated>2026-06-14T06:26:57Z</updated>
<author>
<name>Onur Gündüz</name>
<email>ogunduz@gmail.com</email>
</author>
<published>2026-05-30T09:23:52Z</published>
<link rel='alternate' type='text/html' href='https://git.onurgunduz.com/ffwd/commit/?id=f6e959455add3260a483c89e6fb9aa214d4e6789'/>
<id>urn:sha1:f6e959455add3260a483c89e6fb9aa214d4e6789</id>
<content type='text'>
</content>
</entry>
<entry>
<title>dist: add distributed layer-slice foundation</title>
<updated>2026-06-14T06:26:52Z</updated>
<author>
<name>Onur Gündüz</name>
<email>ogunduz@gmail.com</email>
</author>
<published>2026-05-30T09:11:42Z</published>
<link rel='alternate' type='text/html' href='https://git.onurgunduz.com/ffwd/commit/?id=44a235498ab54ffac97b7a0be659f6da83062d3f'/>
<id>urn:sha1:44a235498ab54ffac97b7a0be659f6da83062d3f</id>
<content type='text'>
</content>
</entry>
<entry>
<title>server: implement HTTP API with libae integration</title>
<updated>2026-06-14T06:18:59Z</updated>
<author>
<name>Onur Gündüz</name>
<email>ogunduz@gmail.com</email>
</author>
<published>2026-05-30T02:07:37Z</published>
<link rel='alternate' type='text/html' href='https://git.onurgunduz.com/ffwd/commit/?id=ab50d908ba2eae14929a07341b1b716cde681c1c'/>
<id>urn:sha1:ab50d908ba2eae14929a07341b1b716cde681c1c</id>
<content type='text'>
</content>
</entry>
<entry>
<title>all: support batched embedding attention</title>
<updated>2026-06-14T06:17:30Z</updated>
<author>
<name>Onur Gündüz</name>
<email>ogunduz@gmail.com</email>
</author>
<published>2026-05-20T03:26:35Z</published>
<link rel='alternate' type='text/html' href='https://git.onurgunduz.com/ffwd/commit/?id=7a88087ba0b5b1b388234f3d2a907673c63bd8cb'/>
<id>urn:sha1:7a88087ba0b5b1b388234f3d2a907673c63bd8cb</id>
<content type='text'>
</content>
</entry>
<entry>
<title>Refactor and enhance pplx_embed and pplx_embed_mlx for improved model support</title>
<updated>2026-03-28T22:10:03Z</updated>
<author>
<name>Onur Gündüz</name>
<email>ogunduz@gmail.com</email>
</author>
<published>2026-03-28T20:27:58Z</published>
<link rel='alternate' type='text/html' href='https://git.onurgunduz.com/ffwd/commit/?id=c47e900b94d18cca1bf600e58f10dbcc2e34c1f7'/>
<id>urn:sha1:c47e900b94d18cca1bf600e58f10dbcc2e34c1f7</id>
<content type='text'>
- Updated `pplx_embed.h` to support new embedding models: `pplx-embed-v1-4b` and `pplx-embed-context-v1-4b`.
- Simplified model configuration structure and constants for better maintainability.
- Enhanced comments for clarity on model architecture and configuration.
- Modified `pplx_embed_mlx.c` to streamline loading of model weights and configuration from `config.json`.
- Improved memory management and error handling in the MLX backend.
- Updated embedding computation to support dynamic hidden sizes based on model configuration.
- Added `pplx_mlx_config` function to retrieve model configuration details.
- Updated README.md
- Added .gitignore
</content>
</entry>
</feed>
