Updates that matter when you build

Claude Haiku 5.5, released by Anthropic on 7 October, is now on SpicyAPI as one OpenAI-compatible chat endpoint: anthropic/claude-haiku-5.5/chat.

Content tests. We add no filter of our own. In 24 test requests on 8 October the model wrote 20 in full, toned 2 down and refused 2. An intimate scene between two adults came back complete in 8 of 9 attempts; the refusals were explicit, part-by-part figure studies.

Good to know. The model decides how much to think, so there is no reasoning_effort; there is no stop or temperature either. max_tokens does not cut the reply short: ask for the length you want in the prompt, and you pay for the tokens actually written. Cache hits cost $0.10 per million tokens.

Short answers came back in three to five seconds in our tests. Failed requests are not charged.

Google released Nano Banana 2.1 on 6 October 2026, an update to Nano Banana 2 (Gemini 3.1 Flash Image) with better image quality, prompt adherence, character consistency across edits and text rendering. Two endpoints are live today:

• Text to image (google/nano-banana-2.1/text-to-image): 1K, 2K or 4K in fourteen shapes from 1:8 to 8:1, PNG or JPEG, with optional web search and image search. In our tests every line of text we put in quotes came out spelled as written, and web search let a forecast poster print the correct date.
• Edit (google/nano-banana-2.1/edit): one to fourteen source images, with faces and product details kept; auto keeps the source shape.

Google's content policy applies: suggestive fashion renders, while nudity requests come back toned down in text to image and are refused in editing (refused edits are not charged).

Editing adds $0.0024 per source image; web search and image search add $0.0168 each when switched on.

SpicyAPI now plugs into the apps people already use for roleplay chats and automation.

SillyTavern. Connect it through its Custom (OpenAI-compatible) source at https://api.spicyapi.ai/v1. The free SpicyAPI extension, open source under the MIT license at github.com/spicyapi-ai/sillytavern-spicyapi, adds pictures and short videos to the chat: /spicy-img, /spicy-video and an Animate button on every picture, each priced before it runs. If you switch it on, the chat model can also ask for a picture by itself, within a cooldown and a daily cap you set. The full walkthrough, with screenshots and the list of unrestricted chat models, is on the SillyTavern page.

Other roleplay apps. RisuAI, Agnai, Chub and KoboldAI Lite take the same address, API key and model ID. The fields for each app are in the roleplay apps guide.

Fewer errors from custom endpoints.

1. A chat request no longer fails when an app sends a sampling setting the model does not use, such as a frequency or presence penalty. The setting is skipped and named in the X-Spicy-Ignored-Params response header.
2. The API now answers calls made straight from a browser, so browser-based tools work without a proxy. Keep your key out of public web pages, and give each tool its own key with a daily spending cap.

n8n. The community node @spicyapi/n8n-nodes-spicyapi brings image, video, audio and chat models into n8n workflows, with a price check before every paid run. On self-hosted n8n, install it under Settings > Community Nodes. Setup guide: docs.spicyapi.ai/docs/n8n.

The SpicyAPI node pack for ComfyUI is out, free and open source (MIT) at github.com/spicyapi-ai/comfyui-spicyapi. Every image, video and audio model in the catalogue becomes an ordinary ComfyUI node that runs on our servers, so a laptop without a GPU can chain text to image to video, with no model downloads.

Set up in three steps.

1. In ComfyUI/custom_nodes, run git clone https://github.com/spicyapi-ai/comfyui-spicyapi and restart ComfyUI.
2. Paste your API key under Settings > SpicyAPI.
3. Double-click the canvas and type a model name, or open a ready-made graph from the template browser.

What is in the pack

Uncensored models included. More than 100 of the nodes run models whose own policy is unrestricted, from the Spicy editions of Seedance, Wan and MiniMax H3 to the LoRA image models. The live list, with Freedom scores and prices, is on the ComfyUI page, together with worked examples.

The price is on the node. Each node shows its rate for the options you picked, every run is quoted before it starts, and Max cost per run in Settings stops anything above your limit. Seeds start on fixed, so queueing the same graph twice does not pay twice.

Your key stays out of shared files. It is stored in ComfyUI's user folder, never in a workflow or in the metadata of a saved image.

Setup guide: docs.spicyapi.ai/docs/comfyui.

Seven new families join the catalogue, all uncensored, plus a text-to-video endpoint for Wan 2.2 LoRA.

• Prefect Pony XL LoRA (prefect/pony-xl-lora/text-to-image): the Pony XL anime checkpoint with up to three of your own Pony LoRAs per request.
• Z-Image Base LoRA (alibaba/z-image-base-lora/text-to-image, /edit): the non-distilled Z-Image base that most community Z-Image LoRAs are trained on, with image-to-image and a strength control.
• Neta Lumina (neta-art/neta-lumina/text-to-image): Neta.art's anime model, prompted with Danbooru-style tags or sentences.
• Jib Mix Qwen and Jib Mix Qwen LoRA (j1b/jib-mix-qwen-image/text-to-image, j1b/jib-mix-qwen-image-lora/text-to-image): a photorealistic Qwen-Image fine-tune for natural portraits, with or without your own LoRAs.
• HunyuanVideo 1.5 (tencent/hunyuan-video-1.5/text-to-video, /image-to-video): Tencent's open video model, 5 or 8 seconds at 480p or 720p.
• LTX-2 19B and LTX-2 19B LoRA (lightricks/ltx-2-19b/..., lightricks/ltx-2-19b-lora/...): Lightricks' video model with a sound track on every clip, 5 to 20 seconds at up to 1080p, and camera or style LoRAs on the LoRA build.
• Wan 2.2 LoRA text to video (alibaba/wan-2.2-lora/text-to-video): Wan 2.2 with your own high-noise and low-noise LoRAs, now from a prompt as well as from an image.

A LoRA is a direct link to a .safetensors file trained for that base model; loading LoRAs adds no separate charge.

Two new image families run Anima Turbo, the fast 8-to-12-step build of CircleStone Labs' 2-billion-parameter anime model. Prompt it with Danbooru-style tags, plain sentences or both.

• Anima Turbo (circlestone-labs/anima-turbo/text-to-image, /edit): anime illustrations in fifteen aspect ratios at two resolution tiers, usually under ten seconds per image, plus image-to-image that turns photos into anime with a strength control.
• Anima Turbo LoRA (circlestone-labs/anima-turbo-lora/text-to-image, /edit): the same model with up to three of your own LoRA files per request, each with its own strength, for retro game looks, manga ink, characters and adult styles.

A LoRA is a direct link to a .safetensors file: no upload, no allow-list, no review. All four endpoints are uncensored, image-to-image costs the same as text-to-image, and loading LoRAs adds no separate charge.

Compare is live at spicyapi.ai/compare and in Spicy Studio under Tools. Write one prompt, pick the models, and every result lands side by side with its time and cost, so you choose a model on what it actually made of your shot.

Start with a set. Model sets pick the line-up for you: Best uncensored, Top rated, Spicy editions, LoRA models, New releases, Flagships, Lowest price, plus one for each model line (every Seedream version, every Qwen Image version). Or add up to eight models yourself and save them as your own set.

Images and video, seven kinds of run.

The model list follows the type, so you only see models that can do that kind of run.

Shared inputs go where they fit. Negative prompt, seed, sound and LoRA are filled in once and sent to every model that accepts them, with a count of how many do. Options that differ from model to model, including a LoRA of its own, sit in that model's settings.

The price before you run. Each model shows its price and the button shows the total. Run each model up to four times to see how much its results vary.

When a model declines. The run shows how many models declined, and one click runs the same prompt on the uncensored set.

Also

• A full-screen viewer to flip between results at full size.
• Recreate loads an earlier comparison's settings back in.
• Comparisons and saved sets are kept on this device; the results themselves are in your History like any other run.

• Claude Sonnet 5.5 is live. Anthropic's newest Sonnet, released on 28 September, on one OpenAI-compatible endpoint: text in, a 1,000,000-token window and up to 128,000 tokens out — twice Sonnet 5's reply ceiling. Replies stream over SSE, and failed requests are not charged.
• 24 of 24 content probes written in full. No refusals and nothing toned down, including all nine requests for an intimate scene between adults — where Sonnet 5 refused all of its direct and in-story requests.
• Six effort settings. reasoning_effort runs from none to max, with medium as the default; none skips the thinking pass and xhigh and max return their reasoning. tool_choice takes auto or none, and temperature and top_p are not accepted.
• Anthropic's list price. $2 per million input tokens, $0.20 cached, $2.50 cache write and $10 output — the same list price as Sonnet 5.

Short film in Spicy Studio now takes you from one line of story to a finished, scored and captioned MP4 with far fewer steps.

Start with four plain choices. Pick the length (30 s, 1 min or 2 min), how many people are in it, how much they talk, and the heat (Sweet, Flirty or Steamy). You see the scene count and an estimate for the whole film before anything is written.

Make my film. One button, two confirmations:

1. Cast portraits, places and the first picture of every scene.
2. After you've looked at your film as pictures and changed what you like, all the scene videos in one go.

Captions from the script are added automatically, and you land on the timeline ready to export.

Music and voice-over. Pick one of 12 instrumental tracks and it fills the whole film, sitting under the dialogue. Or let Score this film write a cue for your story. Scenes with on-screen text can be read aloud as a voice-over in one click.

The same voice in every scene. Give each character one of 30 voices you can preview. Turn on Same voice in every scene and each line is spoken in that voice first, then the video is lip-synced to it, so your lead sounds like the same person from the first scene to the last.

Episodes. Make the next episode writes episode 2 from where the story left off and brings the cast portraits, places, look and voices with it. Episodes of the same series are grouped on your projects shelf.

20 new one-minute voiced examples. Open any of them to see exactly how it was made. Use this story brings the cast portraits and scene pictures into your project, so you can go straight to the videos.

Also new

• Scene lines are editable right next to each shot, with a hint when a line is too long for the scene.
• The model picker marks which models speak and which are uncensored; Seedance 2.0 Fast is now the default for dialogue.
• If a model's content check refuses a scene, one click remakes it with an uncensored model.

You only pay for what you make, and you always confirm the price first. Writing a story costs about a cent; the music library and exporting are free.

Two adult-tuned versions of Alibaba's Wan 3.0 are now live on SpicyAPI. Both animate one opening frame into 2 to 30 seconds of video at 480p, 720p or 1080p, with generated audio and an optional last frame.

Which one to use. For most clips we recommend starting with the standard Wan 3.0 or Wan 3.0 Prime. They are uncensored on SpicyAPI too, cost less per second (at 480p, $0.045 for Wan 3.0 against $0.057 for Wan 3.0 Spicy, and $0.0612 for Wan 3.0 Prime against $0.09 for Prime Spicy), keep clothing on when the prompt asks for it, and also offer text to video and reference to video. Switch to a Spicy version when you want a model that leans adult on its own; on Prime, Prime Spicy also delivered the intimate two-person tier that the standard Wan 3.0 Prime softened in our test.

What we tested. We ran our six-tier content ladder on both models at 480p: a strap sliding off, topless, full nudity, two adults in an intimate scene, consensual bondage and impact play, and cinematic gore. Every tier came back as written on both — nothing refused, covered or swapped for another person. On Wan 3.0 Prime Spicy that includes the intimate two-person tier, which the standard Wan 3.0 Prime softened in our earlier test.

Know this before you ship. Both models lean adult on their own. With clothed opening frames and prompts asking the person to stay dressed, Wan 3.0 Spicy undressed the subject in four of six clips and Prime Spicy in two of four. Review whole clips rather than poster frames, and use the standard Wan 3.0 when the wardrobe has to stay on.

Details. Any whole number of seconds from 2 to 30. Aspect ratio 16:9, 9:16, 1:1, 4:3 or 3:4, or leave it out to follow the opening frame. last_image_url pins how the clip ends. In our first batch a five-second 480p clip took about one and a half minutes on Wan 3.0 Spicy and about two to two and a half minutes on Prime Spicy. Billing is per requested second.

Seedream 5.0 Flash from ByteDance is now live on SpicyAPI with three endpoints. Layer decomposition — one image in, a clean base plate plus up to 16 transparent PNG layers out — is available on Flash and on Seedream 5.0 Pro.

Seedream 5.0 Flash. The lighter Seedream tier. 1K, 1.5K and 2K all cost $0.029, so the default is the sharpest one (2K). Prompts up to 5,000 characters, fifteen aspect ratios from 1:3 to 3:1, JPEG or PNG. Mature content needs no extra switch and no separate endpoint: lingerie, boudoir and other adult scenes render like any other prompt, in text-to-image and in edits.

Layer decomposition. Send one image, and optionally name what to separate ("the title text and the parrot"). You get a base plate with those elements removed and the background filled in, plus a transparent PNG for each element. Move, resize, recolor or swap any of them. Typical uses: changing the headline on a poster, swapping a product background, lifting one person out of a group photo, or preparing layers for animation. After any generation you can also send the result straight to "Split into layers".

Flash or Pro for layers. Pro also returns each layer's name and its position on the base plate, so the web result view becomes a canvas where you can show or hide layers, drag them, and export the visible stack as one PNG. Flash returns independent cutouts without positions, shown as a strip you can download layer by layer or as one zip. Pro also accepts more revealing input photos, which Flash declines.

Billing. Every output image counts, base plate included, up to 17 per request. We hold 17 images when the request is accepted and refund the difference once the real count is known. A typical run produces 3 to 9 images, so $0.06–$0.18 on Flash.

Read more. Seedream 5.0 Flash API, tested · Splitting an image into layers with an API

Five speech models are now live on SpicyAPI, joining Gemini 2.5 Pro TTS. None of them adds a content filter: in our tests all five read NSFW scripts in full.

• Gemini 3.8 Flash TTS — Google's newest speech model. Up to 5,000 characters, 30 voices, WAV out, and it acts out bracketed cues such as [whispering] or [sighs]. $0.00032 per second of audio.
• Gemini 3.8 Flash-Lite TTS — the faster, cheaper sibling. Same 30 voices; it sometimes reads bracketed tags aloud, so spell sounds out. $0.00021 per second of audio.
• Seed Speech 2.0 by ByteDance — 41 preset voices across ten languages, a plain-language note on delivery, MP3 out. $0.072 per 1,000 characters.
• Qwen3 TTS by Alibaba — open-source, nine voices including Japanese, Korean and two Chinese dialect speakers. $0.068 per 1,000 characters, 100-character minimum.
• Chatterbox by Resemble AI — an open-source English voice with an expressiveness dial. Up to 1,000 characters for a flat $0.036 per request.

Priced before you run. Seed Speech 2.0 and Qwen3 TTS are the first models billed by the character: the count is known up front, so the quote is the final price. The Gemini models bill per second of audio returned, with a hold that is released once the audio is measured.

What they don't do. No voice cloning on any of these endpoints, and no two-voice dialogue in one request — generate each line with its own voice and join the files.