Skip to content

The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.

Commits on master

0ff0457
fix QuantizedTensor.is_contiguous (#10956) (#10959)
Urle Sistiana committed 11 months ago
6484ac8
Disable offload stream when torch compile. (#10961)
comfyanonymous committed 11 months ago
f55c98a
Enable async offloading by default on Nvidia. (#10953)
comfyanonymous committed 11 months ago
9d8a817
Update template to 0.7.23 (#10949)
ComfyUI Wiki committed 11 months ago
b59750a
quant ops: Dequantize weight in-place (#10935)
rattus committed 11 months ago
3f382a4
f17251b
block info (#10841)
Haoming committed 11 months ago
c38e7d6
convert nodes_customer_sampler.py to V3 schema (#10206)
Alexander Piskun committed 11 months ago
a2d60aa
d8433c6
Add Z Image to readme. (#10924)
comfyanonymous committed 11 months ago
dd41b74
Fix the CSP offline feature. (#10923)
comfyanonymous committed 11 months ago
55f654d
Merge 3d animation node (#10025)
Terry Jia committed 11 months ago
58c6ed5
Bump frontend to 1.32.9 (#10867)
Christian Byrne committed 11 months ago
234c3dc
1105e0d
add Veo3 First-Last-Frame node (#10878)
Alexander Piskun committed 11 months ago
8938aa3
Add cheap latent preview for flux 2. (#10907)
comfyanonymous committed 11 months ago
f16219e
ComfyUI version v0.3.75
comfyanonymous committed 11 months ago
8402c87
Fix Flux2 reference image mem estimation. (#10905)
comfyanonymous committed 11 months ago
58b8574
ComfyUI v0.3.74
comfyanonymous committed 11 months ago
90b3995
Fix loras not working on mixed fp8. (#10899)
comfyanonymous committed 11 months ago
bdb10a5
Adjustments to Z Image. (#10893)
comfyanonymous committed 11 months ago
0e24dbb
Z Image model. (#10892)
comfyanonymous committed 11 months ago
e9aae31
ComfyUI v0.3.73
comfyanonymous committed 12 months ago
0c18842
Lower vram usage for flux 2 text encoder. (#10887)
comfyanonymous committed 12 months ago
d196a90