Repository navigation
Commit f3c9a1c
committed
fix(opencode): respect model limit.output config instead of capping at 32k
maxOutputTokens() used Math.min(model.limit.output, 32000) which silently
capped any configured output limit at 32k tokens. Now the model's configured
limit.output is honored when set, falling back to 32k only when unset.
Closes anomalyco#293631 parent e5cf3ad commit f3c9a1c
1 file changed
Lines changed: 3 additions & 2 deletions
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
23 | 23 | | |
24 | 24 | | |
25 | 25 | | |
26 | | - | |
| 26 | + | |
27 | 27 | | |
28 | 28 | | |
29 | 29 | | |
| |||
1259 | 1259 | | |
1260 | 1260 | | |
1261 | 1261 | | |
1262 | | - | |
| 1262 | + | |
| 1263 | + | |
1263 | 1264 | | |
1264 | 1265 | | |
1265 | 1266 | | |
| |||
0 commit comments