Closed CleyFaye closed 3 weeks ago
Describe the bug When using the TabbyML VSCode extension, completion suggestion shows a bunch of tags/stuff after the actual completion.
This is for completing TypeScript code using StarCoder2-7B
Information about your version Tabby version: 0.11.1 running on Windows (tabby_x86_64-windows-msvc-cuda122) VSCode extension 1.6.3
Information about your GPU Output of nvidia-smi:
+-----------------------------------------------------------------------------------------+ | NVIDIA-SMI 552.12 Driver Version: 552.12 CUDA Version: 12.4 | |-----------------------------------------+------------------------+----------------------+ | GPU Name TCC/WDDM | Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |=========================================+========================+======================| | 0 NVIDIA GeForce RTX 3080 Ti WDDM | 00000000:2D:00.0 On | N/A | | 0% 51C P8 35W / 370W | 1234MiB / 12288MiB | 9% Default | | | | N/A | +-----------------------------------------+------------------------+----------------------+
Additional context Tabby runs on a remote windows system, while VSCode is on a Linux system. This should not be an issue, but it's worth mentionning.
The command used to run tabbyml is tabby_x86_64-windows-msvc-cuda122.exe serve --model StarCoder2-7B --device cuda
tabby_x86_64-windows-msvc-cuda122.exe serve --model StarCoder2-7B --device cuda
Seems to be a dup of https://github.com/TabbyML/tabby/issues/2277 (fixed in 0.12)
Ah, my bad; I completely missed that release. Sorry.
Describe the bug When using the TabbyML VSCode extension, completion suggestion shows a bunch of tags/stuff after the actual completion.![image](https://github.com/TabbyML/tabby/assets/1065984/1c7b7194-bd6f-47ee-b4b5-b5b71d33696a)
This is for completing TypeScript code using StarCoder2-7B
Information about your version Tabby version: 0.11.1 running on Windows (tabby_x86_64-windows-msvc-cuda122) VSCode extension 1.6.3
Information about your GPU Output of nvidia-smi:
Additional context Tabby runs on a remote windows system, while VSCode is on a Linux system. This should not be an issue, but it's worth mentionning.
The command used to run tabbyml is
tabby_x86_64-windows-msvc-cuda122.exe serve --model StarCoder2-7B --device cuda