---
type: "post"
id: "day-one-comfyui-support-with-bf16-and-int8-7812b2"
slug: "day-one-comfyui-support-with-bf16-and-int8-7812b2"
title: "Day-one ComfyUI support with BF16 and INT8 quantized variants available."
description: "Thesis: A distilled image-model checkpoint becomes more practically usable when a visual workflow tool exposes multiple weight formats and a composable adapter path."
retrieval_nugget: "Thesis: A distilled image-model checkpoint becomes more practically usable when a visual workflow tool exposes multiple weight formats and a composable adapter path."
published_at: "2026-10-13"
updated_at: "2026-10-13"
record_date: "2026-10-13"
date_kind: "published_at"
topics: []
source_urls: ["https://comfyui-wiki.com/en/news/2026-10-09-qwen-image-2-1-turbo"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:day-one-comfyui-support-with-bf16-and-int8-7812b2"
status: "published"
visuals: []
editorial_provenance: {"schema_version":"newruntime-editorial-copy-v1","content_status":"source_grounded_final","final_copy_sha256":"sha256:4919298324bcfd5e4646fe88dd7f133b0645ff86684fb4dc52df55ac32c27679","reviewed_at":"2026-10-13T11:03:37.609Z","source_evidence_count":1,"verified_claim_count":2,"site_analysis_schema_version":"newruntime-site-analysis-v1","site_object_kind":"field_note","observed_fact_count":2,"implication_count":1,"watch_condition_count":1,"related_record_count":0}
analysis: {"schema_version":"newruntime-site-analysis-v1","object_kind":"field_note","thesis":"A distilled image-model checkpoint becomes more practically usable when a visual workflow tool exposes multiple weight formats and a composable adapter path.","observed_facts":[{"text":"The report says Qwen-Image-2.1-Turbo runs generation and editing in eight denoising steps at CFG 1, compared with 40 steps for the base model.","source_urls":["https://comfyui-wiki.com/en/news/2026-10-09-qwen-image-2-1-turbo"]},{"text":"The report lists BF16 and INT8 convrot checkpoints and a Turbo LoRA for ComfyUI; it says the text encoder and VAE are shared with Qwen-Image 2.1.","source_urls":["https://comfyui-wiki.com/en/news/2026-10-09-qwen-image-2-1-turbo"]}],"mechanism":"The report attributes the reduced step count to a Turbo checkpoint with its recommended sampling schedule baked in; users can either swap in the Turbo diffusion model or apply its LoRA to the base transformer in the existing workflow.","why_now":"Comfy-Org’s weights were reported available within hours of the October 9 release, making the accelerated checkpoint usable through an established ComfyUI workflow rather than only a separate pipeline.","implications":["Builders already running Qwen-Image 2.1 can evaluate the Turbo transformer or LoRA without replacing the shared text encoder and VAE."],"evidence_boundary":"The evidence boundary is a ComfyUI Wiki report, not an independent benchmark: it establishes listed formats and setup guidance, but not measured time savings, memory requirements, or comparative output quality.","watch_conditions":["Independent tests should report latency, VRAM use, and image quality across the BF16, INT8, and LoRA paths before treating the acceleration as a net production improvement."],"related_records":[],"new_branch_reason":"No prior New Runtime record was supplied, so this release is opened as a distinct analytical record about Turbo checkpoint deployment in ComfyUI rather than attached to an invented earlier record."}
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/day-one-comfyui-support-with-bf16-and-int8-7812b2/","markdown":"https://newruntime.com/posts/day-one-comfyui-support-with-bf16-and-int8-7812b2.md","json":"https://newruntime.com/posts/day-one-comfyui-support-with-bf16-and-int8-7812b2.json"}
---

# Day-one ComfyUI support with BF16 and INT8 quantized variants available.

## Retrieval answer

Thesis: A distilled image-model checkpoint becomes more practically usable when a visual workflow tool exposes multiple weight formats and a composable adapter path.

Thesis: A distilled image-model checkpoint becomes more practically usable when a visual workflow tool exposes multiple weight formats and a composable adapter path.

The public report says Qwen-Image-2.1-Turbo runs the same 7B generator and editing pipeline in eight denoising steps at CFG 1, rather than the base model’s 40 steps. It lists ComfyUI forms as a BF16 checkpoint, an INT8 convrot quantization, and a Turbo LoRA for use with the base transformer. The text encoder and VAE remain shared with Qwen-Image 2.1.

Mechanism: The report attributes the reduced step count to a Turbo checkpoint with its recommended sampling schedule baked in; users can either swap in the Turbo diffusion model or apply its LoRA to the base transformer in the existing workflow. Why now: Comfy-Org’s weights were reported available within hours of the October 9 release, making the accelerated checkpoint usable through an established ComfyUI workflow rather than only a separate pipeline. The evidence boundary is a ComfyUI Wiki report, not an independent benchmark: it establishes listed formats and setup guidance, but not measured time savings, memory requirements, or comparative output quality.

Implication: Builders already running Qwen-Image 2.1 can evaluate the Turbo transformer or LoRA without replacing the shared text encoder and VAE. Watch condition: Independent tests should report latency, VRAM use, and image quality across the BF16, INT8, and LoRA paths before treating the acceleration as a net production improvement.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)