Default Branch

0758059bd1 · Use Ollama parallelism for startup benchmark · Updated 2026-08-24 10:12:04 +00:00

Branches

1e4fa87437 · Prevent timeout and allow long running connections · Updated 2025-08-28 14:48:57 +00:00    mikhail

147
0
Included

703435d10e · Improve MODEL_SERVER_START_* messages · Updated 2025-08-26 11:42:04 +00:00    mikhail

151
0
Included

eecefd1d52 · for benchmarking, use concurrent requests · Updated 2025-08-11 19:17:09 +00:00    mikhail

168
1
Compare

9773e5f67b · download vast.ai's root certificate in order to make pyworker requests · Updated 2025-07-31 19:47:12 +00:00    mikhail

171
1
Compare

d3be9fe7db · redo metrics tracking for requests, fixes bug wherere some requests were marked as pending, even though they had finished · Updated 2025-07-31 01:56:51 +00:00    mikhail

171
1
Compare

4ac51947b4 · fix pyright errors + revert to old way of handling cancelled api requests · Updated 2025-07-17 22:29:41 +00:00    mikhail

173
1
Compare

6b0f019cf7 · add option to skip auth · Updated 2025-07-15 22:18:09 +00:00    mikhail

176
3
Compare

72a5f6ad13 · update tokenizers deps · Updated 2025-06-11 00:55:25 +00:00    mikhail

179
1
Compare