theroyallab/tabbyAPI
1
0
Fork 0
mirror of https://github.com/theroyallab/tabbyAPI.git synced 2026-04-27 17:51:36 +00:00
Code Issues Packages Projects Releases Wiki Activity

Formatting

Browse Source
This commit is contained in:
turboderp
2025-10-15 10:47:58 +02:00
parent 0af29d957a
commit 486dd0418e
1 changed files with 3 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
backends/exllamav2/model.py
View File

@@ -1338,7 +1338,9 @@ class ExllamaV2Container(BaseModelContainer):
# Automatically set max_tokens to fill up the context # Automatically set max_tokens to fill up the context
max_tokens = unwrap(params.max_tokens, 0) max_tokens = unwrap(params.max_tokens, 0)
if max_tokens <= 0: if max_tokens <= 0:
max_tokens = self.config.max_seq_len - max(context_len, negative_context_len) max_tokens = self.config.max_seq_len - max(
context_len, negative_context_len
)
# Determine if the negative context or the context length is bigger # Determine if the negative context or the context length is bigger
context_to_check = max(negative_context_len, context_len) context_to_check = max(negative_context_len, context_len)