fix(llm): cap auto-detected max_output_tokens when it fills the entir… #10864
| Job | Run time |
|---|---|
| 3m 0s | |
| 2m 40s | |
| 8m 3s | |
| 8m 3s | |
| 9m 41s | |
| 8m 5s | |
| 8m 55s | |
| 33s | |
| 9m 44s | |
| 32s | |
| 28s | |
| 26s | |
| 0s | |
| 0s | |
| 1h 0m 10s |
| Job | Run time |
|---|---|
| 3m 0s | |
| 2m 40s | |
| 8m 3s | |
| 8m 3s | |
| 9m 41s | |
| 8m 5s | |
| 8m 55s | |
| 33s | |
| 9m 44s | |
| 32s | |
| 28s | |
| 26s | |
| 0s | |
| 0s | |
| 1h 0m 10s |