Default Branch

3a7f02acba · fix: use MODEL_LOG env var instead of hardcoded onstart.log to prevent log scanner feedback loop · Updated 2026-07-23 19:18:16 +00:00

Branches

68d8ce4bfd · refactor: use endpoint_id instead of endpoint name for routing · Updated 2025-12-06 22:46:41 +00:00    mikhail

47
1
Compare

222ac2a0dd · default endpoint name · Updated 2025-12-04 18:54:55 +00:00    mikhail

48
0
Included

6b5b1341a7 · update tgi client · Updated 2025-12-04 02:38:42 +00:00    mikhail

54
0
Included

adedb8ba90 · defaults to ENDPOINT_NAME and DEFAULT_MODEL but uses the flag first if present · Updated 2025-12-04 00:57:28 +00:00    mikhail

58
0
Included

0bcd2219ea · Increase model wait time for vLLM · Updated 2025-12-03 20:38:52 +00:00    mikhail

58
0
Included

e143162438 · bumpy pyworker version · Updated 2025-11-26 00:01:23 +00:00    mikhail

62
0
Included

62fbfb061d · more logs · Updated 2025-11-25 02:40:45 +00:00    mikhail

63
3
Compare

9c6ab78503 · Move model log line · Updated 2025-11-24 23:22:23 +00:00    mikhail

69
0
Included

e0449cb3c7 · add llama log · Updated 2025-11-21 18:22:16 +00:00    mikhail

70
0
Included

63550d5af3 · Actual fix -- MOVED TO TEMPLATE · Updated 2025-11-17 18:57:55 +00:00    mikhail

71
4
Compare

74efc2cb42 · bump up version minor number · Updated 2025-11-15 02:07:17 +00:00    mikhail

71
2
Compare

249ca2eb99 · refactor, handle zombie tasks · Updated 2025-11-12 23:23:42 +00:00    mikhail

77
2
Compare

a47c9d1ed0 · remove test bugs · Updated 2025-11-12 02:13:46 +00:00    mikhail

72
0
Included

eedf81c0a3 · Updated readme and .gitignore · Updated 2025-11-12 01:18:40 +00:00    mikhail

74
0
Included

353462ecb8 · try allow parallel requests · Updated 2025-11-11 19:27:05 +00:00    mikhail

77
1
Compare

d63a060202 · Merge pull request #56 from vast-ai/obfuscate-mtoken · Updated 2025-11-10 19:53:17 +00:00    mikhail

78
0
Included

c6521cb6d4 · add ... · Updated 2025-11-07 18:10:35 +00:00    mikhail

79
0
Included

c9d701e8d3 · increase wait time for llm backends · Updated 2025-11-04 00:21:56 +00:00    mikhail

85
1
Compare

b03645d145 · Added model type environment variable so we can actually attempt to benchmark with the right payload. · Updated 2025-10-30 06:25:44 +00:00    mikhail

94
1
Compare

7437028cb2 · Added caller for REPORT_ADDR to backend.py · Updated 2025-10-30 01:02:17 +00:00    mikhail

94
0
Included