huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-11 20:34:54 +00:00
Code Issues Packages Projects Releases Wiki Activity

Adding a few more cards.

Browse Source
This commit is contained in:
Nicolas Patry 2024-11-05 11:24:20 +01:00
parent 5bcb3e6ad2
commit e85dc0a0bb
No known key found for this signature in database
GPG Key ID: D2920555C90F704C
1 changed files with 5 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
launcher/src/main.rs
View File

@ -1498,6 +1498,11 @@ struct ComputeType {
impl ComputeType { impl ComputeType {
fn f16_flop(&self) -> Option<u64> { fn f16_flop(&self) -> Option<u64> {
let card_flop = match &self.card[..] { let card_flop = match &self.card[..] {
// https://www.nvidia.com/en-us/geforce/graphics-cards/40-series/rtx-4090/
// Specs are unclear https://www.itcreations.com/nvidia-gpu/nvidia-geforce-rtx-4090-gpu
"nvidia-4090" => Some(82 * 10u64.pow(12)),
// https://www.nvidia.com/en-us/data-center/tesla-t4/
"nvidia-t4" => Some(65 * 10u64.pow(12)),
// https://www.nvidia.com/en-us/data-center/l4/ // https://www.nvidia.com/en-us/data-center/l4/
"nvidia-l4" => Some(121 * 10u64.pow(12)), "nvidia-l4" => Some(121 * 10u64.pow(12)),
// https://www.nvidia.com/en-us/data-center/products/a10-gpu/ // https://www.nvidia.com/en-us/data-center/products/a10-gpu/