Release 1.21 docs#718
Conversation
Signed-off-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com>
Signed-off-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com>
Signed-off-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com>
Signed-off-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com>
Signed-off-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com>
Signed-off-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com>
Signed-off-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com>
Signed-off-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com>
Signed-off-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com>
- Removing mosaicml/mpt-7b model as it is no longer available in HF - Removing e5-mistral-7b-instruct and stella_en_1.5B_v5 since they are no longer part of release1.21 Signed-off-by: vtirumal <vtirumal@qti.qualcomm.com>
tv-karthikeya
force-pushed
the
release-1.21-docs
branch
from
e799e19 to
0b1e634
Compare
tv-karthikeya
requested review from
ochougul,
quic-amitraj,
quic-hemagnih and
quic-rishinr
as code owners
|
Comments on documentation - I see below as repeated twice- We should remove decode only from GPT-OSS and add Disaggregated serving ready via vLLM Add Gemma3 model also in the list Add a line in the Release notes that if running GPT-OSS models natively via vLLM then we need a PR#685 of the qefficient for python 3.12 Mention the VLM model names for which CB is enabled. Continuous Batching (VLMs): Extended to Vision Language Models with multi-image handling mention "Users can enable the feature by passing “use_onnx_subfunctions=True “ call during export " for onnx subfunctions. Add below also |
quic-hemagnih
left a comment
quic-hemagnih
left a comment
There was a problem hiding this comment.
I see below as repeated twice-
Efficient Transformer Library - 1.21.0 Release Notes
Newly Supported Models
Key Features & Enhancements
Embedding Model Upgrades
Fine-Tuning Support
Efficient Transformer Library - 1.20.0 Release Notes
Newly Supported Models
Key Features & Enhancements
Embedding Model Upgrades
Fine-Tuning Support
We should remove decode only from GPT-OSS and add Disaggregated serving ready via vLLM
GPT-OSS (Decode-Only)
Add Gemma3 model also in the list
Add a line in the Release notes that if running GPT-OSS models natively via vLLM then we need a PR#685 of the qefficient for python 3.12
Mention the VLM model names for which CB is enabled. Continuous Batching (VLMs): Extended to Vision Language Models with multi-image handling
mention "Users can enable the feature by passing “use_onnx_subfunctions=True “ call during export " for onnx subfunctions.
Add below also
Onboarding Guide for adding new Causal models (#574)
Onboarding Guide for adding new Custom ops in QEff (#638)
Organized examples into domain-specific subdirectories (#615)
Signed-off-by: Amit Raj <amitraj@qti.qualcomm.com>
Signed-off-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com> Signed-off-by: vtirumal <vtirumal@qti.qualcomm.com> Signed-off-by: Amit Raj <amitraj@qti.qualcomm.com> Co-authored-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com> Co-authored-by: Amit Raj <amitraj@qti.qualcomm.com>
Signed-off-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com> Signed-off-by: vtirumal <vtirumal@qti.qualcomm.com> Signed-off-by: Amit Raj <amitraj@qti.qualcomm.com> Co-authored-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com> Co-authored-by: Amit Raj <amitraj@qti.qualcomm.com>
Signed-off-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com> Signed-off-by: vtirumal <vtirumal@qti.qualcomm.com> Signed-off-by: Amit Raj <amitraj@qti.qualcomm.com> Co-authored-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com> Co-authored-by: Amit Raj <amitraj@qti.qualcomm.com>
Signed-off-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com> Signed-off-by: vtirumal <vtirumal@qti.qualcomm.com> Signed-off-by: Amit Raj <amitraj@qti.qualcomm.com> Co-authored-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com> Co-authored-by: Amit Raj <amitraj@qti.qualcomm.com> Signed-off-by: Dipankar Sarkar <dipankar@qti.qualcomm.com>
Signed-off-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com> Signed-off-by: vtirumal <vtirumal@qti.qualcomm.com> Signed-off-by: Amit Raj <amitraj@qti.qualcomm.com> Co-authored-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com> Co-authored-by: Amit Raj <amitraj@qti.qualcomm.com> Signed-off-by: Sharvari Medhe <smedhe@qti.qualcomm.com>
Signed-off-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com> Signed-off-by: vtirumal <vtirumal@qti.qualcomm.com> Signed-off-by: Amit Raj <amitraj@qti.qualcomm.com> Co-authored-by: Abukhoyer Shaik <abukhoye@qti.qualcomm.com> Co-authored-by: Amit Raj <amitraj@qti.qualcomm.com> Signed-off-by: Sharvari Medhe <smedhe@qti.qualcomm.com> Signed-off-by: Sharvari Medhe <smedhe@qti.qualcomm.com>
4 participants
No description provided.