1c507a7b12
add qwen3-asr
2026-02-05 00:43:49 +08:00
073723447f
merge main
2026-02-02 21:30:52 +08:00
f2cb6bf976
merge main
2026-02-02 17:56:34 +08:00
901324cd1a
fix qwen3vl-thinking startswith bug
2026-01-31 18:57:42 +08:00
16e3aefaa7
fix aha run multiple inputs bug
2026-01-31 18:37:28 +08:00
292dfd31e4
merge pr/go-restream/20
2026-01-30 22:17:00 +08:00
e3944588fc
index tts stash save
2026-01-30 22:04:23 +08:00
f010087e97
```
...
feat(cli): add direct model inference via new run subcommand
- Add `aha run` CLI subcommand for direct model inference without HTTP service
- Support multiple models including Qwen series, OCR models, ASR models, and voice generation
- Implement input/output handling with file path support and auto-generation
- Add comprehensive documentation in CLI_USAGE.md with examples
- Include performance timing for model loading and inference operations
- Add macOS build target to Makefile with Metal support
```
2026-01-21 18:41:11 +08:00
861c411ba2
feat: CLI Subcommand Support
...
- **CLI Subcommand Support**: Added three new subcommands for better command organization:
- `aha cli` - Download model and start HTTP service (default, backward compatible)
- `aha serv` - Start HTTP service only (requires `--weight-path`)
- `aha download` - Download model only (no service start)
2026-01-20 22:15:45 +08:00
53791efa80
fix modelscope fun-asr-nano-2512 model load error
2026-01-17 13:18:26 +08:00
45b53e41cd
update fmt
2026-01-15 23:20:16 +08:00
f4e95e3937
use linear_b
2026-01-15 23:17:34 +08:00
d9b803d27e
add qwen3 and fun-asr-nano
2026-01-15 21:57:12 +08:00
2c1d5e3a14
update fmt
2026-01-09 18:47:39 +08:00
461eed7556
updata resample use rubato
2026-01-09 16:06:04 +08:00
24194182fc
update get audio data
2026-01-08 19:22:21 +08:00
618d75b263
merge pr/KingBright/13
2026-01-08 00:04:24 +08:00
d7df9afb70
merge metal features
2026-01-07 23:04:00 +08:00
dd70fa8f63
update fmt
2026-01-07 21:49:31 +08:00
6ef8facfde
add glm-asr-nano
2026-01-07 21:46:01 +08:00
981457347a
load audio has problem
2025-12-31 17:39:25 +08:00
3ba4de8a7d
deploy add address
2025-12-26 14:56:17 +08:00
6b8381a765
update
2025-12-26 00:33:07 +08:00
88ee303392
merge pr/guobin211/12
2025-12-26 00:23:14 +08:00
0cbedb24e4
refactor: 优化RMBG2.0图像预处理流程，提前转换RGB数据以减少重复操作
2025-12-25 21:10:05 +08:00
9a4fa636be
add image api and audio api
2025-12-25 20:25:52 +08:00
7c9261567b
refactor: 为RMBG2.0测试方法添加cfg(test)属性，优化extract_image_url函数避免clone
2025-12-25 10:44:32 +08:00
f35631157a
feat: 优化RMBG2.0图像处理性能，支持并行批量处理
2025-12-24 21:42:05 +08:00
06bd7fce04
add RMBGv2.0
2025-12-23 19:23:21 +08:00
675fd6f89e
stash save
2025-12-12 20:40:57 +08:00
855d0cf975
update fmt
2025-12-11 23:30:07 +08:00
d282a0cf40
Merge branch 'voxcpm1.5'
2025-12-11 23:15:05 +08:00
b9947a6516
update voxcpm code to support voxcpm1.5 model
2025-12-11 23:14:43 +08:00
6430338e95
add url crate to handle local path
2025-12-11 19:33:30 +08:00
963924a1f1
stash save
2025-12-11 18:33:35 +08:00
42c95da0e3
refactor: remove unstable Option<&T> branch by returning owned Tensor
2025-12-11 10:50:35 +08:00
0dc2840b0e
update fmt
2025-12-10 16:52:57 +08:00
1a0df13bc7
model test add tps
2025-12-10 16:44:20 +08:00
3eb983b5bb
stash save
2025-12-10 12:57:24 +08:00
515d6da852
replace some function
2025-12-10 00:05:46 +08:00
dc4b593011
add ffmpeg feature
2025-12-09 15:52:26 +08:00
4228616058
add paddleocr-vl
2025-12-09 00:41:30 +08:00
697484cf23
add hunyuan_ocr
2025-12-03 17:21:01 +08:00
7d72cb3baf
use eager_attention_forward
2025-11-27 18:43:16 +08:00
93b4beb316
update fmt
2025-11-27 14:34:15 +08:00
43703d1b63
merge deploy
2025-11-27 14:28:47 +08:00
b51c52fa4c
fix download img blocking error
2025-11-27 14:11:55 +08:00
17f94a93f0
delete print
2025-11-27 13:29:33 +08:00
d1b4f7a1b6
update fmt
2025-11-25 21:54:44 +08:00
de3861b731
update note
2025-11-25 17:45:09 +08:00
jhqxxx
XiaoYang
michaelbguo
jhqxxx
YageGeng