Commits · ece7ffa40a7e167400e57d89b8a73751d095184c · OpenDAS / text-generation-inference

19 Jun, 2023 1 commit
- feat(server): improve flash attention import errors (#465) · ece7ffa4
  OlivierDehaene authored Jun 19, 2023
```
@lewtun, is this enough?

Closes #458
Closes #456
```
  ece7ffa4
05 Jun, 2023 1 commit
- feat(server): batch tokenization for flash causal lm (#411) · 6abec14a
  OlivierDehaene authored Jun 05, 2023
  
  6abec14a
31 May, 2023 1 commit
- feat(server): add retry on download (#384) · 87dc034b
  OlivierDehaene authored May 31, 2023
  
  87dc034b
03 May, 2023 1 commit
- feat(server): support hf endpoint weight layout (#266) · 85aa7e2e
  OlivierDehaene authored May 03, 2023
  
  85aa7e2e
11 Apr, 2023 1 commit

OlivierDehaene authored Apr 11, 2023

OPT models do not all have a `tokenizer.json` file on the hub at the
moment. Can't merge for now.

f26dfd0d

07 Mar, 2023 1 commit
- feat(clients): Python client (#103) · 3fef90d5
  OlivierDehaene authored Mar 07, 2023
  
  3fef90d5
06 Mar, 2023 1 commit
- feat: allow local models (#101) · cd5961b5
  OlivierDehaene authored Mar 06, 2023
```
closes #99
```
  cd5961b5
18 Feb, 2023 1 commit
- feat(server): enable hf-transfer (#76) · 17bc841b
  OlivierDehaene authored Feb 18, 2023
  
  17bc841b
16 Feb, 2023 1 commit
- feat(router): add prometheus metrics scrape endpoint (#71) · 439fcaf8
  OlivierDehaene authored Feb 16, 2023
  
  439fcaf8
15 Feb, 2023 1 commit
- feat(server): improve download logging (#66) · c5a4a1fa
  OlivierDehaene authored Feb 15, 2023
  
  c5a4a1fa
14 Feb, 2023 1 commit
- feat: add safetensors conversion (#63) · 0fbc6919
  OlivierDehaene authored Feb 14, 2023
  
  0fbc6919