Asuka-FM: a T2I DiT Pretrained from Scratch on One RTX 4090
A 206M dual-stream DiT pretrained from scratch on one RTX 4090: about 160 GPU hours, under 55 USD. GenEval 0.609 at 512px, no benchmark…
Text-to-image is a crowded race, and the frontier has moved to editing and control: FLUX.2/3 push open weights and quantization onto consumer GPUs, Qwen-Image and Seedream close in from China, and GPT Image 2 brings reasoning into pixels. This topic tracks model releases, architecture shifts, acceleration work, and the business numbers behind "pretty" becoming "usable".
A 206M dual-stream DiT pretrained from scratch on one RTX 4090: about 160 GPU hours, under 55 USD. GenEval 0.609 at 512px, no benchmark…
HiT adds a learned token-grouping hierarchy to pixel-space diffusion: 1.48 FID on ImageNet 256, JiT-level quality at about 40% of its…
Linum trained a 2.18B pixel-space model P-JiT for $21,038 on 8,415 H100-hours, releasing the weights, code and the training recipe under…
WeChat's WeLM-Image unifies seeing, drawing and editing: 31.7B stored, 5.4B active per token; Turbo renders 1024px in 4.98s on one H20.
LongBanana: expert-verified benchmark, 90 tasks, ~32 reference images each, 4,670 atomic checks. RPR agents lift scores but cut pass rates…
One frozen video model, one LoRA: FrameForge unifies three image tasks. In its table, editing 6.12 beats task-specific; motion 6.84 tops…
IntReal benchmarks unified multimodal models on 100K interleaved samples. NanoBanana2 wins text logic 97.49 versus 75.17, but trails on…
A pretrained autoregressive image model is turned into an any-order generator, beating MaskGIT at inpainting while keeping KV cache and…
SpecSlot treats tokenization as allocation: only each anchor's lowest-frequency slot joins global attention. The paper reports 9.14 FID on…
inclusionAI posted IDM on arXiv, matching the generator's own EMA features onto DMD supervision, no external encoder, 94.25% LongText-CN at…
Alibaba's 9B Qwen-Image-PE rewrites text-to-image prompts, beating GPT-5.5 on the team's own benchmark; sparse prompts make it hurt.
LayerEditor couples instruction editing with reusable RGBA layers in one token model, backed by a 100K paired dataset and two benchmarks.
Bria's Fibo 2.0-dev unifies image editing and text-to-image in one open model, ranked #1 in open-weight editing on Artificial Analysis per…
SenseTime-led MT Forcing curbs drift in autoregressive multi-turn image editing: eight-turn instruction following 8.47 to 9.66 on…
WorldCast tests multi-step world state prediction on 320 cases: GPT-Image-2.5 scores 72.8, Nano Banana Pro 42.8, and the best open model…
OpenMOSS ships a 5B Apache-2.0 DiT image base: bilingual prompts, 4K-token inputs, in-image text rendering, JSON layout control; no…
BFL open-sources FLUX 3 Image Klein, a 5.6B text-to-image and editing model running on 8GB VRAM, first in BFL's internal open-weights…
Microsoft-led LensErase removes any object, text or watermark from one image plus one mask, pretrained on 101M pairs for large gains on…
Qwen ships Qwen-Image-2.1-Turbo: same 7B architecture, denoising cut from 40 steps to 8. APIs cost 0.1 RMB per image; weights stay…
Google's Nano Banana 2.1 hit GA Oct 6 on Gemini 3.6 Flash, halving image prices to 3.36 cents/1K. Benchmarks top Pro; hands-on tests still…
BFL's FLUX 3 Image: 0-1000 grid, caption plus JSON element tables, box-by-box edits leaving other pixels intact, 10 references, native 4K…
arXiv paper MaLiang-Harness names the Program-to-Visual gap — code runs, image wrong — and closes it with a draft-inspect-revise loop for…
Ant Group's RULER replaces scalar SVG rewards with six-item rubrics. MMSVG scores rose from 0.432/0.395 to 0.693/0.683, matching…
Alibaba's Qwen team open-sourced Qwen-Image-2.1 on September 20. The vision generation module weighs in at just 7B parameters, folding…
OpenAI's Images 2.5 cuts image generation latency by up to 50% versus Images 2.0 and adds Sketch: type @Sketch, doodle, and use your…
Tencent Hunyuan paper turns prompts into layered, mouse-editable HTML designs via a coding agent. Open source, Apache 2.0, near 500 GitHub…
Ant's inclusionAI open-sources LLaDA-Image, a 6B model that generates and edits images in one checkpoint; the distilled Turbo runs in 4…
Reverse engineering of Windows Paint and Photos shows that images generated locally by Stable Diffusion on Copilot+ PCs still have their…
Tencent's Hunyuan team released WithEveryone, a framework for identity-preserving group image generation with up to ten reference…
A field study released by INFORMS shows that in a live Instagram campaign for an outdoor-activities company, ads generated by an…
NovelAI ships Diffusion V5 (Curated and Full): over twice the size of V4.5, trained on 268,000 B200 GPU-hours, with a custom 32-channel…
xAI shipped Grok Imagine Image 2.0 on August 7, 2026, rolling it out as the Quality Mode on grok.com/imagine, iOS, and Android, with the…
Alibaba's Qwen-Image-3.0 has officially launched on the Qianwen AI platform, with the Pro and Standard variants opening API access…
Tencent Hunyuan Image-3.0 is fully open-sourced: it uses a unified autoregressive framework for multimodal understanding and image…
Microsoft Research Asia / Microsoft team released Mage-Flow today, a 4B compact text-to-image + instruction-editing unified stack, with…
On July 21, Alibaba Qwen officially launched Qwen-Image-3.0, the third generation of the Qwen-Image image-generation foundation model. If…
JD.com (jd-opensource) officially open-sources JoyAI-Image — a true unified multimodal foundation model. It uses an 8B multimodal LLM…
Reve Lab released its Reve 2.1 text-to-image model on July 9, completing a key iteration just one month after 2.0. The new model recaptures…
ByteDance Seed × University of Hong Kong × Peking University propose SpectraReward in arXiv 2607.11886, using "reading back the prompt" as…
Peking University + Tencent Hunyuan's joint team open-sources GEAR (Guided End-to-End AutoRegression for Image Synthesis), using a dual…
Diffusion-model post-training has long been dragged down by memory and throughput — 12B-level FLUX.1-dev wanting to do customization or RL…
On July 7, 2026, Meta officially launched Muse Image in its official Newsroom — the first native image generation model from Meta's…
EPFL's Lan Feng, Wuyang Li, and Alexandre Alahi, in arXiv 2607.02375v1 released on July 2, let the classic MMD objective function — which…
# Flow-Map GRPO: opens the door to reinforcement learning for deterministic "few-step image generation" Few-step Flow-Map generators — like…
Zheng Xingyu et al. at Beihang University, in arXiv 2607.01642, propose MrFlow — a completely training-free multi-resolution diffusion…
In June 2026, Manchester-based Sourceful released Riverflow 2.5 Pro, built around reasoning image generation. It topped all three Design…
The Qwen team released the Qwen-Image-2.0-RL technical report, porting the RLHF and on-policy distillation (OPD) playbook — matured in LLM…
arXiv 2606.27377 introduces DanceOPD, a unified image-generation framework from ByteDance that addresses the well-known multi-capability…
Hugging Face user Boogu released Boogu-Image-0.1, an open-source multimodal model that unifies "understanding + generation" in a single…
Krea AI released Krea 2, a 12B-parameter DiT (Diffusion Transformer) image generation model. The dual release includes: Krea 2 Base (the…
ByteDance released Seedream 5.0, the next-generation image generation model, following the recent release of Seedance 2.5 (video…
Alibaba Qwen released Qwen-Image-2.0, the next-generation image generation and editing model. The standout: a 7B-parameter model that…
arXiv 2606.19195 introduces Moebius, a 0.2B-parameter image restoration model that matches the quality of 10B-parameter industrial models…
arXiv 2606.13679 introduces InterleaveThinker, a dual-Agent pipeline that adds "interleaved generation" capability to any image generation…
Black Forest Labs released FLUX.2 [klein], a new image generation and editing model available in 4B and 9B variants. The standout: the…
arXiv 2509.25787 introduces EvoQuality, a ByteDance method for training VLMs to assess image quality with zero human annotation. The…
On June 3, Silicon Valley startup Reve released the second-generation image generation model Reve2.0, which on launch day rocketed to #2 on…
Ideogram 4.0 is the company's first open-source-weights model, released on June 3: 9.3B parameters, 34 layers of single-stream Diffusion…
Microsoft's MAI-Image-2.5 image generation model, launched at Build 2026, hit Arena image editing #2 on its first day, putting Nano Banana…
NetEase Youdao released LobsterAI on June 1, a multimodal large-model matrix that is the company's first open-source product to unify image…
Black Forest Labs released the FLUX.2 image generation model family in early May 2026, with 32 billion parameters, runnable directly in…
On April 30, Midjourney officially released version V8.1, the third iteration of the V8 series since its March launch. Compared to V8.0,…
Black Forest Labs has officially released the FLUX.2 image generation model family. Unlike the previous "one model to rule them all"…
Latest data shows that image generation models have become a core driver of AI mobile app growth. According to Appfigures, the download…
On April 21, OpenAI officially released GPT Image 2 (ChatGPT Images 2.0), the successor to DALL-E 3 and the industry's first model that…
Researchers from Tongji University and other institutions have published a paper on arXiv proposing DiT-ST, an innovative framework that…
Alibaba's Tongyi Lab has released Wan2.7-Image, a full-pipeline model that unifies image generation and editing. Unlike the traditional…