LauraGPT
1b1fff2caa
deploy: a078fe3467
2026-09-09 17:33:15 +00:00
LauraGPT
1016e332f6
deploy: 3ea977f55f
2026-09-09 14:19:08 +00:00
LauraGPT
e799871b1e
deploy: 1a310d77df
2026-09-09 11:55:03 +00:00
LauraGPT
2d3e88e3e8
deploy: 130e57a6fd
2026-09-07 22:26:32 +00:00
LauraGPT
2ebbdc2fa7
deploy: 04278203a6
2026-09-07 21:33:12 +00:00
LauraGPT
0acc5d522e
deploy: 50fddb748f
2026-09-07 14:13:24 +00:00
LauraGPT
b8dd9f9c25
deploy: dd79f9dd5a
2026-09-07 13:18:27 +00:00
zhifu gao and zhifu gao
67d63b80a2
deploy: use QwenAudio canonical repository links ( #3413 )
...
Co-authored-by: zhifu gao <[email protected]
2026-07-24 19:01:52 +08:00
LauraGPT
ee9c414166
docs: add community donors page
2026-07-23 02:44:57 +00:00
zhifu gao and LauraGPT
992e3281a1
docs: correct Nano timestamp capability scope ( #3232 )
...
Co-authored-by: LauraGPT <[email protected]
2026-07-15 09:33:38 +08:00
zhifu gao and LauraGPT
d5502d3db1
docs: correct Nano benchmark language scope ( #3230 )
...
Co-authored-by: LauraGPT <[email protected]
2026-07-15 09:00:52 +08:00
zhifu gao
a61ad7ab79
fix(benchmark): correct CER to canonical micro-avg+normalize_zh (SenseVoice 7.81 / Paraformer 10.18 / Fun-ASR-Nano 8.06)
2026-06-23 14:43:19 +08:00
LauraGPT and Claude Opus 4.8
22ccc787f8
Revert Fun-ASR-Nano benchmark row to original (my earlier edit used a different CER method, creating an inconsistent table)
...
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]
2026-06-16 16:41:23 +00:00
LauraGPT and Claude Opus 4.8
cb28b88d8f
docs(benchmark): align Fun-ASR-Nano GPU row to canonical funasr.com numbers
...
The Fun-ASR-Nano single-stream GPU figure contradicted the headline benchmark
on www.funasr.com (which this page is linked from as the full report).
Align to the authoritative source: 17.0x/10.56% -> 21.0x/8.06% (RTF 0.0476).
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]
2026-06-16 09:31:07 +00:00
zhifu gao
0ec430faff
docs: add Korean Benchmark website page
2026-05-26 20:46:27 +08:00
zhifu gao and xiaoyunchong.xyc
8f24081a29
docs: add SEO metadata to documentation pages
...
Co-authored-by: xiaoyunchong.xyc <[email protected]
2026-05-25 14:27:48 +08:00
zhifu gao
2c969da85e
docs: reconcile refreshed GitHub Pages site
...
Publish the newer GitHub Pages surface with vLLM, Agent, Benchmark, and OpenAI-compatible API homepage links.
2026-05-24 21:32:22 +08:00
xiaoyunchong.xyc
3b4d573b7c
feat: add Japanese to all language switchers, create missing ja/ pages
...
- All pages now have English/中文/日本語 consistently
- Created ja/benchmark.html and ja/api.html
2026-05-24 14:57:33 +08:00
xiaoyunchong.xyc
928463e3cc
fix: unify nav bar across all pages (Benchmark link + language switcher)
2026-05-24 14:38:51 +08:00
xiaoyunchong.xyc
4194e5d97d
benchmark: remove specific hardware model references
2026-05-23 22:20:47 +08:00
xiaoyunchong.xyc
4a83dba03b
benchmark: remove CER column (speed-only comparison)
2026-05-23 19:30:09 +08:00
xiaoyunchong.xyc
39203d921d
benchmark: clarify PyTorch backend (vLLM benchmark coming later)
2026-05-23 19:20:09 +08:00
xiaoyunchong.xyc
0cfe67d1e0
benchmark: remove GLM-ASR (community model, speed needs verification)
2026-05-23 19:10:49 +08:00
xiaoyunchong.xyc
7e6b8e8c95
benchmark: add comprehensive 8-model GPU + 3-model CPU comparison
...
- Added Fun-ASR-Nano, GLM-ASR-Nano, Whisper-large-v3-turbo (OpenAI & FunASR)
- Added CPU results section (SenseVoice 17.2x, Paraformer 15.6x, Fun-ASR-Nano 3.6x)
- Added model type badges (NAR/AR/LLM) with architecture explanation
- Updated hero stats and key findings for 8-model comparison
2026-05-23 18:59:18 +08:00
游雁
00884e0c63
Add benchmark page and nav link (preserving all existing content)
2026-05-23 16:57:31 +08:00
游雁
3265136817
Revert "Add benchmark page: FunASR vs Whisper speed & accuracy comparison"
...
This reverts commit 2e2a1f374d
2026-05-23 16:55:18 +08:00
游雁
2e2a1f374d
Add benchmark page: FunASR vs Whisper speed & accuracy comparison
2026-05-23 16:45:17 +08:00
LauraGPT