666 Commits

Author SHA1 Message Date
RapidMark
7948df8ac1
fix(cmake): build HIP backend with PIC so the static-lib PIE link succeeds (#1593) master-666-7948df8 2026-06-02 00:07:48 +08:00
Wagner Bruna
02f06370a7
refactor: call CPU backend functions dynamically (#1591)
Co-authored-by: leejet <leejet714@gmail.com>
master-665-02f0637
2026-06-01 23:41:21 +08:00
stduhpf
f8935d6f25
feat: support img-cfg for edit models (#929)
Co-authored-by: leejet <leejet714@gmail.com>
master-664-f8935d6
2026-06-01 22:54:25 +08:00
stduhpf
be65ac7511
feat: add support for APG (adaptive projected guidance) + unconditionnal SLG (#593) master-663-be65ac7 2026-06-01 00:55:49 +08:00
leejet
20901f6d8e
fix: remove kv padding from flash attention wrapper (#1453) master-662-20901f6 2026-05-31 23:23:19 +08:00
leejet
0982807139
feat: add PiD support (#1585) master-661-0982807 2026-05-31 22:38:39 +08:00
leejet
d2797b8667
fix: correct Gemma3 rope settings and vram limit propagation (#1583) master-660-d2797b8 2026-05-30 22:23:49 +08:00
leejet
d3b2cb047e
fix: split tokens before normalization (#1582) master-659-d3b2cb0 2026-05-30 18:38:46 +08:00
akleine
b4ba55d8d7
fix: prevent crash in case of a mem alloc error and graceful exit (#1566) master-658-b4ba55d 2026-05-30 18:34:07 +08:00
Wagner Bruna
b54bd83a3f
fix: explicitly exclude f8, f64 and i64 tensors from mmap (#1575) master-657-b54bd83 2026-05-30 18:31:08 +08:00
Wagner Bruna
0e4ee04488
fix: correct tae for models that use the flux2 vae (#1571) master-656-0e4ee04 2026-05-28 09:13:16 +08:00
leejet
29ab511fc7
fix: resolve LLM norm tensor names by architecture (#1570) master-655-29ab511 2026-05-28 00:36:16 +08:00
leejet
55c2aed52c
refactor: simplify diffusion model runner params (#1569) 2026-05-28 00:12:35 +08:00
leejet
8eded497e5
fix: preserve frontend tooling in ROCm CI build (#1568) 2026-05-27 21:26:16 +08:00
leejet
92dc7268fc
feat: add microsoft lens support (#1560) master-652-92dc726 2026-05-27 01:04:17 +08:00
schirik
07b2b18e70
fix: skip permission denied errors in recursive_directory_iterator (#1564)
Co-authored-by: Serge F. Chirik <s.chirik@timbel.info>
master-651-07b2b18
2026-05-27 00:56:16 +08:00
leejet
1ceb5bd9df
fix: package ROCm BLAS runtime in Windows artifacts (#1562) master-650-1ceb5bd 2026-05-26 00:57:37 +08:00
leejet
202c6154a2
fix: use flux flow prediction for LTXAV (#1561) master-649-202c615 2026-05-26 00:23:39 +08:00
stduhpf
a397e03488
feat: add Longcat-Image / Longcat-Image-Edit support (#1053)
Co-authored-by: leejet <leejet714@gmail.com>
2026-05-24 02:02:02 +08:00
leejet
72e512a0cc
fix: make macOS binaries use relocatable rpaths (#1552) master-647-72e512a 2026-05-23 12:27:06 +08:00
leejet
0baf721215
feat: add LTX temporal latent upscaler support (#1551) master-646-0baf721 2026-05-23 01:35:13 +08:00
leejet
645e6e9089
feat: add LTX rational latent upscaler (#1549) master-645-645e6e9 2026-05-23 00:28:15 +08:00
stduhpf
cbf92191c3
fix: strip trailing latent channels for preview decode (#1548) master-644-cbf9219 2026-05-23 00:26:40 +08:00
stduhpf
8cf55a3b3b
fix: load TAESD preview-only model correctly (#1547) master-643-8cf55a3 2026-05-23 00:22:35 +08:00
leejet
3a8788cb7d
refactor: unify extra argument parsing (#1540) master-642-3a8788c 2026-05-22 01:00:03 +08:00
leejet
449165caf5
feat: stream LTX VAE temporal tile decoding (#1539) master-641-449165c 2026-05-22 00:25:04 +08:00
stduhpf
adaa599a3b
Feat: Temporal tile custom size with overlap (#1510)
* Temporal tile size + overlap

* add --extra-tiling-args support

---------

Co-authored-by: leejet <leejet714@gmail.com>
master-640-adaa599
2026-05-21 23:44:12 +08:00
leejet
2e3514625a
perf: run LTX audio VAE decode in one ggml graph (#1538) master-639-2e35146 2026-05-21 22:43:14 +08:00
stduhpf
47d8198b69
feat: add taeltx2_3_wide support (#1535) master-638-47d8198 2026-05-21 22:34:12 +08:00
leejet
ef92a0027e
feat: add graph cut markers for LTXAV transformer (#1534) master-637-ef92a00 2026-05-20 23:22:10 +08:00
leejet
b3374e6a71
feat: add LTX spatial latent upscale hires support (#1533) master-636-b3374e6 2026-05-20 22:27:09 +08:00
stduhpf
bdd937f29a
feat: add taeltx2/taeltx2.3 support (#1531) master-635-bdd937f 2026-05-20 22:14:05 +08:00
stduhpf
c51ec7cad9
fix: always load runtimle lora params on runtime backend (#1532) master-634-c51ec7c 2026-05-20 22:13:15 +08:00
leejet
5b0267e941
fix: avoid Vulkan f16 repeat in LTX audio VAE (#1528) master-633-5b0267e 2026-05-19 23:15:26 +08:00
leejet
0045a72b96
fix: trigger ci for docker image changes (#1527) master-632-0045a72 2026-05-19 22:05:03 +08:00
leejet
99bd062546
fix: update sycl docker image to oneapi 2025.3 (#1526) 2026-05-19 21:59:15 +08:00
leejet
9d8c9e4279
fix: build web UI for Windows ROCm server releases (#1525) 2026-05-19 21:53:48 +08:00
George Sofianos
caa823a8c0
ci: add RDNA1 + RDNA2 targets for ROCm 7.13 (#1511) 2026-05-19 01:38:02 +08:00
leejet
22c8c40b0d
sync: update ggml (#1520) 2026-05-19 01:30:11 +08:00
leejet
b706d682ad
fix: restore singleton dims for LLM outputs (#1518) 2026-05-18 23:47:10 +08:00
leejet
b758b7de13
fix: only enable TAE after successful load (#1517) 2026-05-18 23:32:03 +08:00
Wagner Bruna
f683c88a28
feat: make negative max_vram control the amount of spare vram (#1503) master-625-f683c88 2026-05-18 23:00:06 +08:00
Christoph
21fd4e6788
ci: add CUDA Docker image support for NVIDIA Spark GB10 (#1512) master-624-21fd4e6 2026-05-18 22:52:01 +08:00
leejet
830804262b docs: update news 2026-05-18 00:24:29 +08:00
leejet
82e03ef137 ci: add inactive pr clean up workflow 2026-05-18 00:09:45 +08:00
leejet
baf7eda1e4
refactor: minify vocab files (#1509) master-621-baf7eda 2026-05-17 23:06:58 +08:00
Wagner Bruna
e7eb92fd84
feat: add Gradient Estimation sampler (#1484) master-620-e7eb92f 2026-05-17 22:54:28 +08:00
leejet
50134e51dd
refactor: split guidance composition (#1506) master-619-50134e5 2026-05-17 20:20:16 +08:00
leejet
e43b24cf48
feat: add ltx2.3 flf2v support (#1505) master-618-e43b24c 2026-05-17 18:40:14 +08:00
stduhpf
06accf2b39
feat: add ltxav latent2rgb projection matrix (#1502) master-617-06accf2 2026-05-17 17:52:05 +08:00