Default Branch

56f783b38d · Update. · Updated 2025-10-21 19:29:48 +03:00

Branches

191553a298 · super secret update · Updated 2025-10-15 14:49:49 +03:00    alihan

1
25

ad3bc6ceaa · downgrade grpcio, grpcio-tools to 1.70.0 · Updated 2025-03-22 01:23:11 +03:00    alihan

3
0
Included

30c3f58a00 · downgrade grpc to 1.67.0. waiting for fix bd8f8a86e0 · Updated 2025-03-01 00:25:11 +03:00    alihan

17
0
Included

ee095766d9 · handle -gzip suffix in etag for integrity check fixes #633 · Updated 2025-02-26 01:08:15 +03:00    alihan

19
0
Included

1e0a7bbb9e · workaround amd device capabilities · Updated 2025-02-24 20:45:23 +03:00    alihan

20
2

f9a1e5342b · update notice in README · Updated 2025-02-18 14:41:09 +03:00    alihan

24
0
Included

477e3a5e4c · make max_parallel_downloads configurable, increase download chunk size to 8MB · Updated 2025-02-15 00:26:41 +03:00    alihan

31
0
Included

928214d479 · apt-get debian noninteractive in circleci · Updated 2025-02-08 05:40:51 +03:00    alihan

35
0
Included

c5b7b599a0 · tailscale expires ordering · Updated 2025-02-08 03:39:42 +03:00    alihan

42
3

d17fdd223d · await ensure_downloads_dir in brackets · Updated 2025-02-06 14:34:39 +03:00    alihan

42
1

5157d80a46 · remove tenacity dependency, implement simple retry logic instead · Updated 2025-02-04 00:56:38 +03:00    alihan

42
0
Included

72329ba984 · patch for manual discovery, set known_peers · Updated 2025-02-02 02:06:57 +03:00    alihan

43
0
Included

51b5c2ca9b · add model downloading section to README · Updated 2025-02-01 23:23:05 +03:00    alihan

55
0
Included

0bebf8dfde · fix indent · Updated 2025-01-30 23:21:28 +03:00    alihan

58
0
Included

e6b4f2993c · fix prompt output spacing in tui · Updated 2025-01-29 22:01:30 +03:00    alihan

62
0
Included

4887be5103 · parallelise model loading · Updated 2025-01-29 05:32:59 +03:00    alihan

68
0
Included

141de0d011 · increase chatgpt api response timeout to 900 seconds · Updated 2025-01-29 05:03:00 +03:00    alihan

69
0
Included

9c1bea97e8 · fix embed_tokens for last layer in qwen models · Updated 2025-01-29 02:09:45 +03:00    alihan

76
0
Included

7c649085a1 · fix eta/speed for resuming an existing download, using the session downloaded bytes · Updated 2025-01-27 22:23:18 +03:00    alihan

80
0
Included

a3766f538a · add exception for mlx-community/DeepSeek-R1-3bit and mlx-community/DeepSeek-V3-3bit in tokenizers test · Updated 2025-01-26 23:32:48 +03:00    alihan

96
0
Included