One stack for all AI. Run any model, at the lowest price.
- Tweets 1,172
- Following 415
- Followers 7,734
- Likes 2,801
FFmpeg tools are now available on Runware 🛠️
Run common FFmpeg commands over your files through the same Runware API as every generation model, with nothing to install or run locally.
- Upload up to 10 video, audio, or image inputs per task
- Raw filter graphs, stream maps, and input/output options
- Output to MP4, WEBM, MOV, MKV, MP3, WAV, AAC, FLAC, OGG, PNG, JPG, WEBP
- Trim, scale, overlay, extract audio, grab frames, and more
from $0.02 per GB-second
3
2
9
1,815
@Kling_ai's Kling 4.0 is live on Runware on D0 🎬
Up to 30 seconds of 1080p video with native audio, plus:
- Multi-keyframe support, including first, middle, and last frames
- Up to 15 inputs per request across keyframes, reference images, reference videos, and elements
- Prompt-based editing of existing clips
From $0.133 per second of output, with 15% off regular pricing for the next 7 days.
It joins the 19 Kling models already on our API, across video, image, avatar, and lip-sync.
Already calling Kling models on Runware? Switch your model ID to klingai:kling-video@4.0
2
3
16
7,316
MiniMax M3.1 Flash from @MiniMax_AI is now on Runware 🧠
- Text, image & video input
- 1M context, up to 524K output tokens
- Adjustable reasoning depth
- 42 on the Artificial Analysis Intelligence Index
- 72.5% AutomationBench-AA, 73.7% DeepSWE
- Cheaper per token than MiniMax M3
3
3
21
9,616
FLUX 3 Image Klein is now on the Runware API ⚡
@bfl_ai's FLUX 3 image model, in two variants. Both handle text-to-image and editing with up to 4 reference images, plus bounding boxes that place, move, resize, or remove elements by region.
- FLUX 3 Image Klein (bfl:flux@3-image-klein): distilled, 16 steps by default, from $0.006 per 1024×1024 image
- FLUX 3 Image Klein Base (bfl:flux@3-image-klein-base): undistilled, 50 steps by default, adds CFG scale and negative prompts, from $0.021 per 1024×1024 image
Live now on the Runware API
4
1
16
1,825
Claude Haiku 5.5 from @AnthropicAI is now on Runware
- Anthropic's fastest, lowest-cost Claude model
- First Haiku with adjustable effort: trade cost for depth
- 1M context, 128K max output, text and image input
- Built for subagents, extraction, classification, and support
From $0.10/1M input and $0.50/1M output tokens
3
1
11
1,996
It's your last chance to get 50% off FLUX 3 Image on Runware!
FLUX 3 Image is now live on Runware 🖼️
The first image model built on @bfl_ai 's new FLUX 3 multimodal backbone, the same architecture behind FLUX 3 Video.
- Native 4K generation and editing
- Multi-reference composition with up to 10 images, preserving identities and fine details
- Bounding boxes to place, move, resize, or remove elements
- Targeted precise editing
- In-place text editing and multilingual typography
Launch week: 50% off through Oct 8
- 0.75K: $0.020/image (regularly $0.041)
- 1K: $0.024/image (regularly $0.048)
- 2K: $0.050/image (regularly $0.100)
- 4K: $0.303/image (regularly $0.607)
Live in the Playground and on the API: bfl:flux@3-image
4
3
10
1,027
Nano Banana 2.1 is on Runware 🍌
- Sharper composition and subject consistency
- Up to 14 reference images and 10 reference videos
- Up to 4K output across 14 aspect ratios
- 3 thinking levels
Meet Nano Banana 2.1, our latest image generation and editing model 🍌
This upgraded version outperforms our previous models across the board, with notable leaps in visual design, mask-based editing, and subject consistency to help you create more natural-looking images.
3
2
16
9,573
Qwen-Image-2.1 Pro is on Runware 🎨
Alibaba's unified model for professional quality generation and editing.
Text rendering and typography, supporting Chinese and English
Native 2K, reshaped to any aspect from square to 8:1 ultrawide
Prompt rewriting before generation, direct or agent mode
Optional thinking mode for higher-quality outputs
$0.075/image
4
4
18
1,199
Grok Imagine 1.5 Lite is on Runware 🎬
xAI's faster, cheaper distill of Imagine Video 1.5
- Text-to-video + image-to-video with native audio
- 480p, 720p, 1080p, up to 15s
- 5s image-to-video in ~16-32s, vs ~42s on 1.0
- ~8x less GPU time than 1.0 at 720p
From $0.02/s
15
3
41
1,777
FLUX 3 Image is now live on Runware 🖼️
The first image model built on @bfl_ai 's new FLUX 3 multimodal backbone, the same architecture behind FLUX 3 Video.
- Native 4K generation and editing
- Multi-reference composition with up to 10 images, preserving identities and fine details
- Bounding boxes to place, move, resize, or remove elements
- Targeted precise editing
- In-place text editing and multilingual typography
Launch week: 50% off through Oct 8
- 0.75K: $0.020/image (regularly $0.041)
- 1K: $0.024/image (regularly $0.048)
- 2K: $0.050/image (regularly $0.100)
- 4K: $0.303/image (regularly $0.607)
Live in the Playground and on the API: bfl:flux@3-image
4
11
70
10,415
Runware co-founder Ioana Hreninciuc is heading to @TechCrunch Disrupt 2026.
On Tuesday, October 13, Ioana will take the stage to present Competing for AI Talent: Pay, Equity, and Retention, unpacking how founders are rethinking compensation, culture, and team building in the race for AI talent.
If you’ll be at Disrupt, come say hi!
#TechCrunchDisrupt
12
14
565
Recraft V4.1 Flash is now on Runware ⚡
@recraftai's new flash-class model
- 14 aspect ratio presets, up to 1536×768
- Steer output with a target color palette or a fixed background color
- Strong on portraits, complex scenes, and logo concepts
- Recraft rates prompt adherence above comparable fast models
- Pricing, $0.007 per image
Available now on the Runware Playground and API.
1
1
10
2,134
Seedream 5.0 Flash is live on Runware 🎨
ByteDance's fast 5.0 model variant for high-volume image editing work.
- Edit with up to 10 reference images
- Decompose an image into up to 16 PNG layers
- Target local edits with precise prompt coordinates
From $0.019 /output image
2
7
1,611
Jev and Laya are now live on Runware 🎯
Two “System One” decision-making models: instead of generating text, they return a choice, a score, or a yes/no probability which your app can act on directly.
- Jev, $0.042 / 1M input tokens
- Laya, FREE for the first two weeks, until October 12, then $0.02 / 1M input tokens
Available on Runware’s /v1/systemone API endpoint now.
3
2
10
494
We’re excited to sponsor @a16z speedrun’s AI Faire at @SFTechWeek on October 9.
Runware founder Ioana will be presenting, alongside live demos from AI startups across the speedrun community.
Request a spot! partiful.com/e/WrF0Hv7cZLnp5…
1
6
559
We're at @HumanXCo [X] Amsterdam! Runware is here as a sponsor, with founder Ioana taking to the stage to talk about unifying fragmented customer data and scaling AI safely, including how to give AI systems useful customer context without compromising privacy.
Attending? Come say hello!
Day one recap 🇳🇱
The first HumanX in Europe opened last night at RAI Amsterdam, with three conversations worth the trip on their own.
#HumanX #EuropeanAI #AI
2
1
11
519
P-Video-2 Pro is live on Runware. H3-based, 480p and 768p, speed and quality modes, first/last frame control, and native audio. From $0.01/s this week only!
Today, we launch P-Video-2-Pro, based on MiniMax H3: The best quality. The best speed.
Text or image in. Video and audio out.
• 𝗜𝗻𝗽𝘂𝘁: Text or first-frame image, with optional last-frame conditioning
• 𝗦𝗽𝗲𝗲𝗱: ~2.0s for a 5s clip at 480p · ~4.3s at 768p in Speed mode
• 𝗣𝗿𝗶𝗰𝗶𝗻𝗴: From $0.02/s Speed · $0.04/s Quality
• 𝗠𝗼𝗱𝗲𝘀: Speed or Quality
• 𝗖𝗼𝗻𝘁𝗿𝗼𝗹: First and last frame, seed, aspect ratio, and prompt upsampling with Off, Turbo, or Max
Available via our inference partners @replicate, @Cloudflare, @eachlabs, @scenario_gg, @TellersAI, @wavespeed_ai, @wiroai, @runware, and @togethercompute
Validated by our benchmark partners @datapointai, @RapidataAI, and @DesignArena. Benchmarks to be released soon 👀
👉 Free Playground: playground.pruna.ai/
📚 Model docs: docs.api.pruna.ai/guides/mod…
🧩 API: dashboard.pruna.ai/login
3
4
22
2,056
FLUX Video Edit [fast] is on Runware 🎬
Edit objects, settings, characters, text on screen, and dialogue from a text prompt, with @bfl_ai's new FLUX Video Edit [fast]; 50 seconds to edit a 10-second clip at 720p. Available now on Runware’s One API.
Pricing from 0.03$/s
2
4
15
6,079
P-Video-2 from @PrunaAI is now live on Runware 🎬
Pruna AI’s latest T2V, I2V, and audio-conditioned video model is here on Runware, delivering up to 20s outputs at 1080p, 48fps. Draft mode at half the per-second rate for fast iteration.
From $0.025/s full · $0.015/s draft
1
2
19
980
Two new MiniMax H3 models, Max Turbo and Fast, are live on Runware 🎬
MiniMax H3 ranks top three for video editing and image-to-video on Artificial Analysis, generating up to 15 seconds with native audio.
- Max Turbo runs the same resolutions as Max at half the price per second.
- Fast drops to 480p for quick iteration when you don't need full resolution.
75% off both through Sep 14 – from $0.01 per second.
1
6
1,879
MiniMax H3 Max now available on Runware 🎬
High-throughput variant of MiniMax H3, post-trained for tighter prompt following and polished audio-visual results. Ranked #1 on Design Arena's i2v Leaderboard (Elo 1,349).
Live on Runware’s enterprise API, from $0.05/second
ALT https://www.designarena.ai/leaderboard/image-to-video - H3 Max leading Design Arena Image to Video Leaderboard.
4
2
7
3,293
Gemini Omni Flash 1.1 is now on Runware 🎬
- Output at 360p, 720p, 1080p, up to 4K
- Extend clips in 3-10s increments up to 30s total
- Define start and end frames for controlled interpolation
- Pass reference videos to preserve character and visual style
Price from $0.036/s
7
1
15
814
Recraft V4 Styles is now on Runware🎨
Style-consistent generation from reference images without the need for training, finetuning, or LoRAs. Four models covering raster and vector output, from standard to pro fidelity.
- Recraft V4 Styles: raster, branded content and campaigns
- Recraft V4 Styles Pro: higher-fidelity raster for editorial and premium output
- Recraft V4 Styles Vector: SVG, scalable design assets
- Recraft V4 Styles Pro Vector: premium SVG for brand systems and icon libraries
8
3
28
8,291
Wan 3.0 Prime is now on Runware 🎬
Following yesterday's Wan 3.0 launch, Prime is now available on Runware. Keeping the same API shape, multimodal workflows, output quality and impressive 30 second outputs, while introducing faster inference for rapid generation.
Wan 3.0 is now on Runware on D0 🎬
- Text-to-video, frame, and reference modes in one model
- 30-second clips
- Up to 10 reference images, 5 reference videos, 5 reference audio inputs
- Thinking mode for complex multi-input prompts
- Designed for product demos, branded storytelling, and music-led sequences
10
6
25
2,391
Wan 3.0 is now on Runware on D0 🎬
- Text-to-video, frame, and reference modes in one model
- 30-second clips
- Up to 10 reference images, 5 reference videos, 5 reference audio inputs
- Thinking mode for complex multi-input prompts
- Designed for product demos, branded storytelling, and music-led sequences
17
3
53
18,087
Two new Ideogram models are now on Runware 🎨
@ideogram_ai 4.0q 9.3B, open-weight model:
- Quantized V4.0, running on Runware’s hardware
- Bounding-box layout control for precise text and element placement
- Accepts structured JSON prompting for repeatable output
- Color-palette guidance & multilingual text rendering
- LoRA support for controllable inference
Ideogram Background Remover:
- Pass any image, get back a clean transparent PNG cutout
- Strong edge fidelity for hard boundaries such as typography & logos
3
1
29
3,002
LTX-2.5 Fast and pro from @ltx_io are on Runware on D0!
2 variants: Fast for speed and low cost, Pro for higher fidelity.
- Text-to-video, image-to-video, and audio-to-video
- Up to 4K resolution, and 1080p for Pro
- Clips up to 20 seconds at 720p/1080p
- Native audio generation
2
3
10
2,645
Seedance 2.5 is here. And it has a range of new features and capabilities.
Our technical writer, @felixsanzm, has written 7 guides to help you get the best results. Thread 🧵
1
6
15
2,039
Seedance 2.5 from @BytePlusGlobal is coming soon on Runware.
runware.ai/seedance-2-5
2
1
7
846
Open-source video models have trailed closed-source by a wide margin. H3 just closed that gap
MiniMax H3 is now the SOTA open video generation model in both @arena and @ArtificialAnlys benchmarks.
7
2
15
1,246
Today we're launching Sonic Inference Pods, our network of modular data centers built for AI inference.
Each one delivers 1MW of compute from a 20-foot shipping container, deployed wherever there's power.
Learn more: bit.ly/4hNdg9I
21
22
106
71,695
MiniMax H3 is now on Runware 🎬
- New open-weights flagship video model from MiniMax
- Native synced audio, generated in the same pass as the video
- Multi-reference conditioning keeps subjects and scene identity consistent across shots
- Extend existing clips or audio segments
8
1
166
23,824
LoRA training, live on Runware ⚙️
- Five base models: FLUX.2 [klein] 4B & 9B, FLUX.1 [dev], Qwen-Image, Z-Image Base
- Train, then immediately run inference without leaving the Runware ecosystem
- Safetensors output: auto-imported to your account, or available for download
- Private by default, you own the output
1
2
15
746
BytePlus Video Enhancement is now on Runware 🎬
- Upscale to 4K at 120fps
- Frame interpolation and detail enhancement
- Denoise, deblur, and SDR-to-HDR
Standard mode for all content types. Pro for real-people footage, producing noticeably better skin and fine detail.
4
3
15
5,721
Introducing Runware Skills 🧩
Ask your agent for the outcome. Skills route to the right model. As new models ship, the routing updates automatically.
- 30+ packaged skills
- Works in Claude Code, Cursor, any MCP agent
- Nothing to host or install: clone the repo and go
12
7
24
3,364
Seedream 5.0 Pro by @byteplusglobal is now on Runware 🎨
- Up to 10 reference images per generation for swaps & composites
- Set exact dimensions, or describe the shape in your prompt for auto aspect ratio
- Output in 1K or 2K, PNG or JPEG
1
2
17
1,342
Runware is now live in the Dify Marketplace 🚀
@dify_ai is a popular platform for building AI agents and workflows. Now you can plug Runware straight in as a model provider.
With a Runware API key and zero custom integration work, drop Runware into any Dify chatbot, agent, or workflow node.
2
1
9
731
NightCafe Select is live, and entries close July 12.
We're proud to have Sarah Impey from our team on the judging panel alongside Marco Pittarello from PROMPT Magazine and community artist Gully Ahmed. They're looking for craft, concept, cohesion, and intent across a full body of work.
One winner walks away with a $1,000 grant and a feature in Prompt Magazine.
Open to all skill levels, all creators, worldwide. If you make AI art, don't sit this one out.
Enter here: eu1.hubs.ly/H0wDtkn0
1
3
7
838
Seed Audio 1 from ByteDance is now on Runware 🎧
- Dialogue, SFX, music, & ambient sound from a single prompt
- Multi-speaker conversations with distinct character voices
- Voice cloning from up to 3 reference audio clips
- Character voice derived from a portrait image
1
3
13
6,527
Runware’s MCP is live 🤖
Any AI tool - Claude Code, VS Copilot, Cursor, Codex - can now call the full capabilities of Runware, just by chatting with your agent.
- Install to Claude Desktop in seconds
- Every model, every parameter, always up to date
- Generate images, video, audio, 3D, text, remove backgrounds, upscale, caption
- Works in Claude Code & other popular MCP-compatible tools
Runware just entered the agentic world
10
6
20
1,948
We teamed up with @NightCafeStudio for an AI art competition for all skill levels.
Submit a project that tells a story start to finish. Get reviewed by expert judges and compete for a $1,000 grant + a feature in Prompt Magazine.
Ready to tell your story? eu1.hubs.ly/H0wmHgH0
3
6
625
HappyHorse 1.1 guides are live on our docs:
- Multi-character casting: lock real reference images into every cut
runware.ai/docs/models/aliba…
- Cinematic storyboarding: multi-shot reels from a single prompt
runware.ai/docs/models/aliba…
Everyone has a horse living inside them.
It is imagination — the ideas that haven't had their chance to run free yet.
The painter's horse waits inside the canvas.
The director's horse waits behind the lens.
The worker's horse waits in the corner of their desk.
They never left. They were just silenced by endless revisions. Caged by rules and frameworks.
Give creation back to the creators. Let the horse race through the lens, crash into the canvas, run wild between the studio and the workspace. Nobody can stop a happy horse. The joy of creating has always belonged to you.
HappyHorse 1.1 Is Now Available, launching alongside the HorsePower AI Cinema Awards.
co-created short film "Are You Happy", presented by HappyHorse CPP creators @Shanzyin_ai
3
2
15
8,791
An open model can do the task for $1.20/M output tokens. The closed one charges $30. And on Friday, a government switched off two of Anthropic's newest models for every customer worldwide.
OSS models belong in your stack.
Read why: runware.ai/blog/the-closed-s…
3
4
10
572
Ray 3.2 from @LumaLabsAI is now on Runware 🎬
- Text-to-video + image-to-video with start and end frame
- 5s or 10s output, up to 1080p, multiple aspect ratios
- Native HDR and EXR export for professional color-grading
- Looping and video extension supported
2
2
15
3,894
Runware now offers free managed hosting for OpenClaw🦞 the fastest-growing open-source AI agent framework.
- No subscription, fully pay-as-you-go only for the tokens your agents consume
- Deploy from your Runware dashboard in minutes
- All frontier models available
8
5
29
6,226
