Comments (3)
Hey @hustxiayang, IIRC, applying the tree-attention results in significantly more compute than using argmax, so the throughput cost is too high to justify the value.
Don't quote me on this but IIRC without the tree the speedup is 2x, while with the tree there is a maximum of ~3x speedup but throughput significantly decreases (divided by 10-20x).
To confirm with @Narsil eventually, hope it sheds some light on the question in the meantime.
from text-generation-inference.
Hi, I suppose you observe speedups from one of:
text-generation-inference/gemma-7b-it-medusa
text-generation-inference/Mixtral-8x7B-Instruct-v0.1-medusa
text-generation-inference/Mistral-7B-Instruct-v0.2-medusa
If that is the case, which dataset(s) did you use?
Another question: do you plan to support typical sampling mentioned in their paper?
from text-generation-inference.
This issue is stale because it has been open 30 days with no activity. Remove stale label or comment or this will be closed in 5 days.
from text-generation-inference.
Related Issues (20)
- DeepSeek-Coder-V2-Lite-Instruct-AWQ HOT 1
- RuntimeError: weight lm_head.weight does not exist HOT 1
- Latest Docker image fails while initializing gemma2 HOT 3
- Server crashes when provided with invalid JSON schema HOT 2
- Meta-Llama-3.1-405B-Instruct-BNB-NF4 - Shard process was signaled to shutdown with signal 11 HOT 1
- paged attention in tgi is fake? HOT 2
- OpenAI style endpoint does not take in chat template automatically HOT 2
- FinishReason enum not compatible with OAI api HOT 3
- AttributeError: 'NoneType' object has no attribute 'replace' HOT 2
- Can't run llama3.1-70b at full context HOT 15
- how to use the model's checkpoint in local fold? HOT 1
- Fixes #3908 HOT 2
- Tools still giving EoF errors on generated JSON HOT 4
- LLama 3/3.1 70B Outputting "!!!!!!"; Shorter Context HOT 5
- Request failed during generation: Server error: Batch ID 408 not found in cache. HOT 5
- Regex response type is not respected HOT 2
- Possible config mismatch between TGI and transformers (`hidden_act` vs. `hidden_activation`) HOT 4
- Feature request: Add documentation and examples for adding additional API endpoints. HOT 3
- local installation fail HOT 7
- 墙内用户如何不使用梯子运行docker容器.Chinese mainland users must use a proxy software when running Docker. How can they avoid using a proxy software? HOT 5
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from text-generation-inference.