One stack for all AI. Run any model, at the lowest price.
- Tweets 1,078
- Following 416
- Followers 7,280
- Likes 2,678
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
20
23
107
70,933
Draft mode is live for Seedance 2.5 on Runware
settings.draft returns a fast 480p preview, perfect for quick iterations.
Pass the returned draftId back as inputs.draftId to output your clip in full, native, 1080p.
๐๐ง๐ญ๐ซ๐จ๐๐ฎ๐๐ข๐ง๐ ๐๐ซ๐๐๐ญ ๐๐จ๐๐ ๐๐จ๐ซ ๐๐ซ๐๐๐ฆ๐ข๐ง๐ ๐๐๐๐๐๐ง๐๐ ๐.๐.
Generate in 480p to explore ideas faster and at lower cost. Once you find the shot you want, generate the final in native 1080p โ while keeping a high degree of consistency in composition, motion, and overall creative direction.
๐๐ซ๐๐๐ญ ๐๐๐ฌ๐ญ. ๐๐ข๐๐ค ๐ฒ๐จ๐ฎ๐ซ ๐ฌ๐ก๐จ๐ญ. ๐
๐ข๐ง๐ข๐ฌ๐ก ๐ข๐ง ๐๐๐๐๐ฉ.
5
2
14
1,158
Weโre excited to sponsor @a16z speedrunโs 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/WrF0Hv7cZLnp5โฆ
1
5
455
We're proud to be partnered with @ComfyUI as an inference provider for the launch of Comfy Router; think OpenRouter, for generative media: a frictionless way to build across the multimodal ecosystem, supported by the speed and cost efficiency of Runware's API.
Comfy Router is live
One API for frontier image, video, 3D, and audio models.
Same model string. Same arguments. No new SDK, no new key, no redeploy.
What Comfy Router gives you:
โ Explicit routing. You name the provider, we call that provider. It's down? The request fails there. No silent fallback.
โ Every job returns the provider that ran it. Log it, bill it, debug it.
โ Async. submit() returns a request ID immediately. The queue retries 429s and transient errors until a slot opens. subscribe() submits and polls to completion.
โ Batch-friendly. Queue a few hundred jobs, hold the IDs, pull results as they land. Nothing blocking on a 5-min video render.
โ 24h retention on inputs and outputs, then deleted.
โ Comfy credits. No sub, no Router fee.
Providers at launch: Comfy. Runware, Wavespeed, Fal, Higgsfield. Multi-provider where the model supports it.
Get Your API Key with the link below. โฌ๏ธ
14
2
43
2,291
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 ๐ณ๐ฑ
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
10
460
๐ We've added deep-dive guides for @PrunaAI's excellent, H3-based, P-Video-2 Pro! If you're just jumping into P-Video-2 Pro, start here:
- Writing prompts for full shots and sequences:
runware.ai/docs/models/prunaโฆ
- Camera moves, physics and full-body motion:
runware.ai/docs/models/prunaโฆ
- Dialogue, lip sync and native audio:
runware.ai/docs/models/prunaโฆ
- Starting and ending on frames you choose:
runware.ai/docs/models/prunaโฆ
- Speed, quality, size and length:
runware.ai/docs/models/prunaโฆ
1
3
18
1,658
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.
โข ๐๐ป๐ฝ๐๐: Text or first-frame image, with optional last-frame conditioning
โข ๐ฆ๐ฝ๐ฒ๐ฒ๐ฑ: ~2.0s for a 5s clip at 480p ยท ~4.3s at 768p in Speed mode
โข ๐ฃ๐ฟ๐ถ๐ฐ๐ถ๐ป๐ด: From $0.02/s Speed ยท $0.04/s Quality
โข ๐ ๐ผ๐ฑ๐ฒ๐: Speed or Quality
โข ๐๐ผ๐ป๐๐ฟ๐ผ๐น: 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 ๐
๐ Free Playground: playground.pruna.ai/
๐ Model docs: docs.api.pruna.ai/guides/modโฆ
๐งฉ API: dashboard.pruna.ai/login
3
4
23
1,991
@deepseek_ai's V4.1 Flash, now available on Runware, from $0.15 per 1M input tokens. Drop model ID deepseek:v4.1@flash into our OpenAI-compatible chat completions endpoint to get started.
๐ Introducing DeepSeek-V4.1-Flash: smarter, faster, more efficient.
๐น Introducing the smallest model in our new architecture family, with native visual understanding.
๐น Designed for greater capability, faster inference, higher throughput, and scaling to larger models.
1/6
1
1
9
984
Launch DeepSeek V4.1 Flash: runware.ai/models/deepseek-vโฆ
OpenAI-compatible endpoint docs: runware.ai/docs/platform/opeโฆ
2
168
FLUX Video Edit [fast]
โ Editing video: runware.ai/docs/models/bfl-fโฆ
โ Removing objects: runware.ai/docs/models/bfl-fโฆ
โ Replacing and adding elements: runware.ai/docs/models/bfl-fโฆ
โ Changing the setting: runware.ai/docs/models/bfl-fโฆ
โ Restyling a clip: runware.ai/docs/models/bfl-fโฆ
โ Dialogue, on-screen text and localization: runware.ai/docs/models/bfl-fโฆ
โ Preparing source video: runware.ai/docs/models/bfl-fโฆ
1
1
276
P-Video-2
โ Writing prompts for scene, motion and sound: runware.ai/docs/models/prunaโฆ
โ Resolution, aspect ratio, frame rate and length: runware.ai/docs/models/prunaโฆ
โ Animating a still with a first frame: runware.ai/docs/models/prunaโฆ
โ Pinning the first and last frame: runware.ai/docs/models/prunaโฆ
โ Native audio, dialogue and lip sync: runware.ai/docs/models/prunaโฆ
โ Driving a clip from your own audio: runware.ai/docs/models/prunaโฆ
โ Draft mode and reproducible reruns: runware.ai/docs/models/prunaโฆ
1
83
