Skip to content

Pull requests: city96/ComfyUI-GGUF

New pull request New
38 Open 36 Closed
38 Open 36 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

MiniMax H3 mmproj support for Qwen3 VL 32B GGUFs
#473 opened Aug 5, 2026 by Nif00 Loading…
Cache device-side constants in dequant kernels
#472 opened Aug 3, 2026 by lanslote Loading…
1
Fix QK-norm .scale/.weight key mismatch on Flux-compat GGUF loads (silent NaN/black image)
#470 opened Aug 2, 2026 by ayanoby Loading…
2 of 3 tasks
Ensure force_patch_weights is set when partially loading/unloading models
#469 opened Jul 26, 2026 by blepping Contributor Loading…
feat(ops): add GGMLTensor.dequantize() for generic core cast path
#468 opened Jul 23, 2026 by Log-Dog012 Loading…
3
Fix loading of BF16 bare nn.Parameters (LTX-2 learnable_registers)
#467 opened Jul 16, 2026 by hc20k Loading…
Handle WeightAdapters in ops.move_patch_to_device
#461 opened Jun 16, 2026 by blepping Contributor Loading…
Add Ideogram 4 architecture support
#460 opened Jun 12, 2026 by DanrisiUA Loading…
5
Add Ideogram support and improve BF16 dequantization handling
#459 opened Jun 9, 2026 by molbal Loading…
5
Report a float compute dtype for quantized GGMLTensor
#456 opened Jun 7, 2026 by fxd0h Loading…
Add Ideogram-4 diffusion model architecture support
#455 opened Jun 7, 2026 by fxd0h Loading…
nodes: skip self.model.to(device_to) walk in unpatch_model on mmap'd weights
#445 opened May 10, 2026 by Booyaka101 Loading…
1
4
Add mistral3 architecture support for Ministral 3B
#440 opened Apr 18, 2026 by insecure-erasure Loading…
1
2
added qwen3.5 support
#438 opened Apr 15, 2026 by TheJoshCode Loading…
Add basic Ministral 3 3B support (Ernie Image)
#436 opened Apr 14, 2026 by deinferno Loading…
7
Update loader.py, fix typo
#434 opened Apr 6, 2026 by RubenGarcia Loading…
feat: add torch dequantization for IQ1_S, IQ1_M, IQ2_XXS, IQ2_S, IQ3_XXS, IQ3_S
#433 opened Apr 6, 2026 by octo-patch Loading…
1
3
Dynamic VRAM support
#427 opened Mar 5, 2026 by rattus128 Contributor • Draft
36
Enhance loader compatibility for sdcpp exports
#425 opened Mar 4, 2026 by rmatif Loading…
GGUF loader fixes: tokenizer assignment, embed cast call, and path resolution
#412 opened Jan 16, 2026 by radicazz Loading…
Fix lumina2 pad token shape mismatch for some GGUF conversions
#392 opened Dec 24, 2025 by vaclavmuller Loading…
9
use cublas_ops for a 15-20% speedup on matmuls with some loss of quality.
#389 opened Dec 21, 2025 by Ph0rk0z • Draft
Adjust Qwen Image quant type
#370 opened Nov 27, 2025 by YarvixPA Contributor Loading…
Initial attempt at gemma2 tokenizer
#358 opened Nov 8, 2025 by city96 Owner • Draft
Update requirements.txt
#353 opened Oct 25, 2025 by wzgrx Loading…
ProTip! no:milestone will show everything without a milestone.