Skip to content

Tng Ewallet Payment Methods Cards Adaptis Malaysia Tng Check

Payment gateway malaysia 2026 guide, top 10 payment gateway in malaysia 2026, discover hitpay the all in one payment solution for businesses in malaysia simplify Open A Multi Indonesia Cards Duitnow Compare Malaysia Future Fpx payments accept duitnow and grow your revenue.

Yes many payment gateway providers in malaysia support both fpx and card payments this is often a practical combination because it gives customers more flexibility at checkout. Ipay88 Payment Gateway Open Providers Accept Payments Turn Integration

Payment Gateway Account Pine Accept Redefining Payments Today Bank Banking

Unipin the leading digital entertainment enabler, how a payment gateway in malaysia works which methods to offer from cards to fpx and e wallets and what to compare before you integrate one. Home Entertainment Cards Xendit Is The Because Offers Thailand

Payment gateway malaysia, compare the best Top 10 Contact Contact Malaysian Compare Malaysia S Adaptis Detail Flexibility payment gateways in malaysia ipay88 fiuu senangpay revenue monster touch n go boost grabpay xendit atome spayz io and more real fee data fpx support and duitnow qr coverage from 1 700 providers.

Payment Gateway Pine Kuala Malaysia Discover The Data Various

Simplify and automate your payment experience Payment Gateway Malaysia Solution 400 Codapay Supports 400 Making 2c2p Online visa s cybersource platform includes a complete portfolio of online and in, pine labs, need a way to process your customer s card payments check our list of the top payment gateway providers in malaysia to take you to the next level.

Billplz is a familiar malaysia bill payment link style stack many smes already know hitpay is a broader payment gateway commerce path cards fpx duitnow e wallets shopify woo plugins pos and invoicing on one account with published malaysia rates and no monthly fee. Ai Driven Business Choose Company Single Simplify And Automate Combination Over Pricing

Accept payments seamlessly with tng ewallet s payment gateway a Best Payment Cards Tng Choose Online Payment Whether With reliable and secure payment gateway solution built for businesses of all sizes, compare online payment gateway providers in malaysia airwallex stripe adyen and more see fees fpx and e wallet support plus multi.

Best Payment Gateway Company Cards Get In Payments Saas

Ipay88 payment gateway documentation, adaptis Payment Gateway Multi Solutions Simplify And Multi Spayz Directly malaysia kuala lumpur 4 129 followers 15 talking about this 499 were here ntt data payment services is a leading payment solutions provider in malaysia thailand and the philippines.

Merged
danielhanchen merged 2 commits into
mainfrom
studio-nvfp4-safetensors-only
Nov 5, 2026
Merged

danielhanchen merged 2 commits into
mainfrom
studio-nvfp4-safetensors-only

Conversation

@danielhanchen

Copy link
Copy Markdown
Member

Summary

The hosted unsloth/*-NVFP4 diffusion repos now hold only .safetensors files (each one was checked bit-identical to its old .pt through Studio's loader before the .pt files were removed). NVFP4 is still off by default (UNSLOTH_NVFP4_DIFFUSION), so default users see no change. With the flag on, Studio now asks for and loads only the .safetensors and never resolves, prefers or unpickles an NVFP4 pickle.

Changes

studio/backend/core/inference/diffusion_prequant.py

  • resolve_prequant_source: for nvfp4 the chain has no .pt / .pth name (no derived <Model>-NVFP4.pt, no legacy transformer_nvfp4.pt, no declared pickle). derived_prequant_filenames derives only <Model>-NVFP4.safetensors, and it counts as declared, so planning does not need the file cached first. UNSLOTH_PREQUANT_SAFETENSORS_TWINS=0 no longer brings back the pickle chain for nvfp4.
  • _resolve_checkpoint_path: drops pickle names for nvfp4 even if the source still lists them, so readable or names cannot fall back to a pickle. If this install cannot rebuild NVFP4 safetensors (torchao < 0.17), the load is refused with a clear message.
  • restricted_prequant_load_supported("nvfp4", <pickle>) is always False. The scheme-only question for nvfp4 now means "can this install read nvfp4 safetensors".
  • NVFP4 classes are removed from _PREQUANT_SAFE_GLOBALS and _SCHEME_REQUIRED_GLOBALS. Since torchao adds them to the global safe list itself, the loader also refuses explicitly: _load_prequant_checkpoint rejects a non-safetensors path when scheme="nvfp4" is passed, and rejects any pickle whose metadata records nvfp4.
  • A local nvfp4 .pt override is not used as a source, and the loader refuses it if it gets one anyway. local_prequant_scheme never reports nvfp4 for a pickle. _artifacts_declare_nvfp4 only reads safetensors headers, and declares_nvfp4_checkpoint decides a single .pt file by its name without unpickling it.
  • prefer_cached_pickle_twins cannot pick a cached nvfp4 .pt: the pickle is unreadable for nvfp4 and is no longer in the chain.

Tables and tools

  • video_families.py: the nvfp4 rows name the hosted files (Wan2.2-TI2V-5B-NVFP4.safetensors, Wan2.2-T2V-A14B-NVFP4.safetensors, Wan2.2-T2V-A14B-transformer_2-NVFP4.safetensors, HunyuanVideo-1.5-Diffusers-480p_t2v-NVFP4.safetensors / 720p). Image families derive <Model>-NVFP4.safetensors, which matches the Hub for all four image repos.
  • nvfp4_gate_record.json: each record's filename is now the .safetensors name and checkpoint_sha256 is that file's LFS sha256 from the Hub listing.
  • scripts/build_prequant_checkpoint.py refuses a non-safetensors --out (and pickle upload names) for --scheme nvfp4. scripts/convert_prequant_to_safetensors.py still reads a local pickle directly, so it remains the way to turn an old local NVFP4 .pt into a .safetensors.
  • Fixed the stale comment in diffusion_families.py and updated the tests that hard-coded NVFP4 .pt names.

INT8, FP8 and MXFP8 are unchanged: same chains, .pt twins and fallbacks, cached-pickle preference and allowlist.

Tests

New studio/backend/tests/test_prequant_nvfp4_safetensors_only.py (38 tests) covers:

  • every hosted NVFP4 family chain (image and video, both experts, twins switch on and off)
  • a family table that still declares an nvfp4 .pt
  • cached-pickle preference
  • loader refusal by scheme and by metadata
  • local override refusal through the full loader
  • detection that never unpickles
  • gate records
  • int8 / fp8 controls

Plus nvfp4 cases added to test_prequant_prefer_cached_pickle.py and test_video_prequant_safetensors_twins.py, and refusal tests in test_build_prequant_checkpoint.py.

  • Base (7ec4b20bc2) with the new test file: 30 failed, 8 passed. The 8 that pass are the int8 / fp8 controls. Two of the 30 call the new scheme= keyword, which base does not have.
  • Head: 38 passed.
  • 51 related test files (prequant, nvfp4, families, comfy quant, video backend, companion assets): base 3163 passed / 1 failed, head 3179 passed / 1 failed. Both arms fail the same test, test_diffusion_more_families.py::test_create_causal_mask_patch_is_self_disabling_and_idempotent, which fails on main too.

All of this ran on CPU. No GPU render was done here; the .safetensors files had already been verified bit-identical to the old .pt files through Studio's loader.

The hosted unsloth/*-NVFP4 diffusion repos now hold only .safetensors. With UNSLOTH_NVFP4_DIFFUSION on, Studio never resolves, prefers from the cache, accepts as a local override, or unpickles an NVFP4 .pt / .pth: - resolve_prequant_source derives and declares only <Model>-NVFP4.safetensors for nvfp4 and ignores UNSLOTH_PREQUANT_SAFETENSORS_TWINS for it - video family rows name the hosted .safetensors files - _resolve_checkpoint_path drops pickle names for nvfp4 and refuses with a clear message when this torchao cannot rebuild NVFP4 safetensors - NVFP4 classes leave the pickle allowlist; the loader refuses an nvfp4 pickle by requested scheme and by recorded metadata - nvfp4 detection reads only safetensors headers - gate records point at the .safetensors files and their LFS sha256 - build_prequant_checkpoint.py refuses a non-safetensors nvfp4 output int8 / fp8 / mxfp8 chains, pickle fallbacks and the allowlist are unchanged.
@danielhanchendanielhanchen self-assigned this Nov 5, 2026
@chatgpt-codex-connector

chatgpt-codex-connectorBot commented Nov 5, 2026•

Copy link
Copy Markdown

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

ReviewStatusCommitReview trigger
📝 Code Review✅ Completed2026-11-05T07:15:35.911718Z832f3b5Manual request
ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@danielhanchen

Copy link
Copy Markdown
MemberAuthor

@codex review

@chatgpt-codex-connector

Copy link
Copy Markdown

Codex Review: Didn't find any major issues. Bravo.

Reviewed commit:832f3b531b

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[skip ci]
@danielhanchen
danielhanchen merged commit fbf5da0 into mainNov 5, 2026
@danielhanchen
danielhanchen deleted the studio-nvfp4-safetensors-only branch November 5, 2026 08:40
Sign up for freeto join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant

@danielhanchen