Nano Banana 2.1 (gemini-nano-banana-2.1), GA on October 6, 2026, is Google's newest and recommended default image model, updating Nano Banana 2 with better text rendering and character consistency, 1K to 4K output and up to 14 reference images.
Faster of OpenAI's two September 2026 GPT Image 2.5 models, for everyday generation and editing; OpenAI cites up to 50% lower latency than Images 2.0. Quality settings run from low to max; alias of the 2026-09-08 snapshot.
Dated September 8, 2026 snapshot that the gpt-image-2.5-flare alias currently resolves to, for callers who want Flare's output fixed if the alias moves. Billed at the same token rates as GPT Image 2.
OpenAI's most capable image model as of October 2026 (released September 2026), aimed at precise, detail-heavy generation and editing at the cost of longer generation times; quality from low to max. Alias of the 2026-09-08 snapshot.
Pinned September 8, 2026 snapshot of GPT Image 2.5 Sunburst and the default behind its alias; text and image in, image out, through the generations, edits and Batch endpoints.
xAI's current image model, released August 7, 2026 and recommended in its docs for image work. Adds a quality setting (low, medium, auto), editing with up to five source images, 1k or 2k output and 21:9 and 5:2 ratios.
Nano Banana 2 Lite (gemini-3.1-flash-lite-image), GA since June 30, 2026, is the fastest and cheapest Gemini image model, targeting sub-2-second latency at 1K resolution only. Google names it the migration path for the original Nano Banana.
Nano Banana Pro (gemini-3-pro-image), Google's Pro-tier image model, GA since May 2026 after a November 2025 preview. Built for complex layouts with accurate text, Search grounding and up to 4K output; Nano Banana 2.1 is the newer default.
Nano Banana 2 (gemini-3.1-flash-image), the Flash-tier image model previewed in February 2026 and GA in May, outputs 512px to 4K and accepts video as context. Deprecated October 6, 2026 for Nano Banana 2.1, with no shutdown date set.
OpenAI's April 2026 image generation and editing model, with flexible output sizes and high-fidelity image inputs; alias of the gpt-image-2-2026-04-21 snapshot. Superseded in September 2026 by GPT Image 2.5 Sunburst and Flare.
Pinned April 21, 2026 snapshot behind the gpt-image-2 alias, launched with ChatGPT Images 2.0. Supports inpainting through the edits endpoint and, since August 2026, a transparent-background preview.
Higher-quality Imagine image tier above Image 1.0, resolving to the 20260403 build. xAI retires it on November 2, 2026, after which requests are served by Grok Imagine Image 2.0 at low quality.
Dated April 3, 2026 build of Grok Imagine Image Quality, which xAI lists as an alias of that model for pinned output. The model behind it retires on November 2, 2026, with Grok Imagine Image 2.0 as the replacement.
Mutable alias that tracks the newest Grok Imagine Image Quality build, still the 20260403 version as of October 2026. Its target retires on November 2, 2026; xAI points users to Grok Imagine Image 2.0 instead.
Dated March 2, 2026 snapshot of Grok Imagine Image 1.0, the build the grok-imagine-image base name currently points to; pin it to keep output unchanged. Superseded by Grok Imagine Image 2.0.
Base slug for Grok Imagine Image 1.0, xAI's lowest-cost image tier, for text-to-image generation and image editing. It currently resolves to the 2026-03-02 snapshot; Grok Imagine Image 2.0 is the newer recommended model.
FLUX.2 [klein] 4B (January 2026), BFL's smallest and fastest image model: a 4B distilled rectified-flow transformer that does text-to-image and multi-reference editing in under a second, runs on consumer GPUs, Apache 2.0.
FLUX.2 [klein] 9B (January 2026): the larger klein model, trading a little speed for higher image quality in unified generation and multi-reference editing; BFL positions it as the best quality-to-latency option, under a non-commercial licence.
Z.ai's main image model since January 2026, pairing a 9B autoregressive generator with a 7B diffusion decoder. Open-weight (MIT) and aimed at accurate in-image text for posters, slides and diagrams, at 512 to 2048 px.
Alias OpenAI added in December 2025, alongside GPT Image 1.5, for the image model used in ChatGPT; OpenAI now labels it the previous ChatGPT image model. Deprecated, with API removal on 2026-12-01 in favour of GPT Image 2.5.
