Update README.md
Browse files
README.md
CHANGED
|
@@ -13,21 +13,6 @@ We run diffusion and video models fast, and we do it on other people's weights.
|
|
| 13 |
Most of what is in this org is a by-product of that work, published because it
|
| 14 |
is more useful outside our cluster than inside it.
|
| 15 |
|
| 16 |
-
## Quantized weights
|
| 17 |
-
|
| 18 |
-
FP8 (e4m3), INT8, NF4 and precision-cast variants of open models — FLUX.1-dev,
|
| 19 |
-
Qwen-Image, Qwen-Image-Edit, Wan 2.1, MAGI-1, LongCat-Video-Avatar, Ovi,
|
| 20 |
-
MiniCPM-V. These are the exact artifacts our fleet loads.
|
| 21 |
-
|
| 22 |
-
Every card states which tensors were actually changed and which were left in
|
| 23 |
-
their original dtype, because "quantized" on its own tells you nothing about
|
| 24 |
-
whether the text encoder came along for the ride. Each one is linked to its
|
| 25 |
-
upstream repository through `base_model`, and inherits the upstream license —
|
| 26 |
-
the FLUX derivatives are non-commercial, the Qwen and Wan derivatives are
|
| 27 |
-
Apache-2.0.
|
| 28 |
-
|
| 29 |
-
→ [Quantized weights we publish](https://huggingface.co/collections/wavespeed/quantized-weights-we-publish-6a86dfeb8b351020868c4950)
|
| 30 |
-
|
| 31 |
## Demo Spaces
|
| 32 |
|
| 33 |
Bring-your-own-key demos for the models we host, and static reference pages for
|
|
@@ -37,16 +22,4 @@ account rather than ours.
|
|
| 37 |
|
| 38 |
→ [Text to image](https://huggingface.co/collections/wavespeed/text-to-image-demos-6a86dfefb74b9952f2f0ede1)
|
| 39 |
· [Image editing and enhancement](https://huggingface.co/collections/wavespeed/image-editing-and-enhancement-demos-6a86dff23faf95a0968db9f1)
|
| 40 |
-
· [Video generation](https://huggingface.co/collections/wavespeed/video-generation-demos-6a86dff5c9fc2d33fc7e5fe4)
|
| 41 |
-
|
| 42 |
-
## Credit upstream
|
| 43 |
-
|
| 44 |
-
Nothing here is a model we trained. The collection below points at the
|
| 45 |
-
repositories whose weights actually run behind our endpoints — start there if
|
| 46 |
-
you would rather run them yourself.
|
| 47 |
-
|
| 48 |
-
→ [Open weights behind our endpoints](https://huggingface.co/collections/wavespeed/open-weights-behind-our-endpoints-6a86dffa3593fe268469c128)
|
| 49 |
-
|
| 50 |
-
---
|
| 51 |
-
|
| 52 |
-
Hosted inference and docs: [wavespeed.ai](https://wavespeed.ai)
|
|
|
|
| 13 |
Most of what is in this org is a by-product of that work, published because it
|
| 14 |
is more useful outside our cluster than inside it.
|
| 15 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
## Demo Spaces
|
| 17 |
|
| 18 |
Bring-your-own-key demos for the models we host, and static reference pages for
|
|
|
|
| 22 |
|
| 23 |
→ [Text to image](https://huggingface.co/collections/wavespeed/text-to-image-demos-6a86dfefb74b9952f2f0ede1)
|
| 24 |
· [Image editing and enhancement](https://huggingface.co/collections/wavespeed/image-editing-and-enhancement-demos-6a86dff23faf95a0968db9f1)
|
| 25 |
+
· [Video generation](https://huggingface.co/collections/wavespeed/video-generation-demos-6a86dff5c9fc2d33fc7e5fe4)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|