Default Branch

303be9304c · docs: improve accuracy of LLM library docs (#12530) · Updated 2025-10-08 07:21:07 +08:00

Branches

de3e0e7d3c · clean up, but no longer working with tool calls? · Updated 2025-10-08 07:11:47 +08:00

30
14

f40f665596 · add comments, cleanup · Updated 2025-10-08 06:42:50 +08:00

23
11

c87b910232 · WIP: stable ordering for tool args · Updated 2025-10-08 06:38:58 +08:00

2
1

df23ca2307 · ollamarunner: measure only active time · Updated 2025-10-08 05:48:18 +08:00

6
2

06b7ee7781 · discover: Disable flash attention for Jetson Xavier (CC 7.2) · Updated 2025-10-08 05:09:01 +08:00

2
1

b72fd226a7 · update shifting logic · Updated 2025-10-07 14:57:19 +08:00

2
6

f580006d10 · thinking: force newer qwen3 models to always use thinking · Updated 2025-10-07 13:40:01 +08:00

2
1

7f06e96ef8 · convert: slice gate_up weight · Updated 2025-10-07 07:29:51 +08:00

6
2

b91c1f6749 · update tests · Updated 2025-10-04 05:49:49 +08:00

23
4

f944382424 · lint · Updated 2025-09-30 11:10:38 +08:00

30
3

abc6a300de · model: tweak renderer for qwen3coder · Updated 2025-09-29 06:41:13 +08:00

30
1

c5cd7fbead · works for 3.1, but regression in 3??? · Updated 2025-09-27 05:35:06 +08:00

36
4

909232168d · deepseek tests · Updated 2025-09-24 05:08:17 +08:00

42
1

ffaf2e7916 · update tests · Updated 2025-09-23 05:25:51 +08:00

52
3

4ef2b2852d · server: serve original error for remote models · Updated 2025-09-21 07:46:32 +08:00

49
1

220a0da37e · simplify expand path · Updated 2025-09-20 04:12:23 +08:00

52
1

b47b9d9063 · s/From*Slice/From*s/ · Updated 2025-09-17 00:50:59 +08:00

70
1

c10a40db99 · parser: tidy up parameter/message parsing · Updated 2025-09-16 09:09:05 +08:00

72
1

92f77a32fc · gemma3: make embedding non-causal · Updated 2025-09-16 06:25:23 +08:00

74
1

7eb0ff7dca · set_rows · Updated 2025-09-16 04:01:18 +08:00

75
1