Aion-3.0-Mini now available on OpenRouter (131k context, $0.70/M in, $1.40/M out)
TL;DR
AionLabs releases Aion-3.0-Mini on OpenRouter. Built on the DeepSeek family, the model supports 131k context at $0.70 per million input tokens and $1.40 per million output tokens.
What changed
Aion-3.0-Mini is now live on OpenRouter with 131k context. Vibe Builders and Basic Users gain direct access to the multi-model roleplaying system built on the DeepSeek family. Developers can call it through the standard OpenRouter endpoint for collaborative generation workflows.
Specs
- •Context window 131k tokens
- •Pricing input $0.70 per M tokens
- •Pricing output $1.40 per M tokens
- •Model ID aion-labs/aion-3.0-mini
- •Vendor docs https://openrouter.ai/aion-labs/aion-3.0-mini
Why it matters
Vibe Builders running extended storytelling sessions now have a 131k context option at $0.70 per million input tokens, undercutting several other roleplay-focused entries on the same platform. Basic Users can start multi-model collaborative chats without managing separate DeepSeek instances. Developers gain a single API path that routes through specialized sub-models for narrative tasks.
What to watch for
Compare output consistency against DeepSeek-R1 on OpenRouter during long roleplay threads. Developers should test context retention on 100k-plus token transcripts and monitor per-turn latency under the listed pricing.
Who this matters for
- Vibe Builders: Use the 131k context window to maintain deep narrative consistency in long-form roleplay sessions.
- Developers: Integrate the OpenRouter endpoint to access multi-model collaborative generation for story apps.
Harsh’s take
Aion-3.0-Mini represents the continued commoditization of specialized fine-tunes. By wrapping the DeepSeek architecture into a multi-model collaborative workflow, AionLabs is targeting the narrative niche where standard instruct models often fail or become repetitive. The pricing is aggressive: at $0.70 per million input tokens, it undercuts many proprietary roleplay models while offering a massive 131k context window.
Operators should look past the roleplay branding. This is a test case for whether collaborative sub-model routing actually produces better long-context coherence than a single large model. If the latency remains low, this architecture proves that specialized model ensembles are the most cost-effective way to handle complex, creative logic.
Watch the OpenRouter benchmarks to see if this ensemble approach holds up against DeepSeek-V3 raw outputs.
by Harsh Desai
About DeepSeek
View the full DeepSeek page →All DeepSeek updatesGo deeper
More AI news
- Daily RoundupGPT-5.6 Sol price drop, DeepSeek V4 Flash Vision, and image models on Replicate
OpenAI cut GPT-5.6 Sol rates on Vercel while DeepSeek added vision to its Flash model and Alibaba released new image generators on Replicate.
- Weekly DigestCursor iPad and Origin beta, Claude Code subagent forking, and Codex CLI 0.149 updates you can run today
Cursor added iPad support and Google Workspace plugins, Claude Code enabled default subagent forking with GitLab tools, and OpenAI Codex shipped new CLI versions plus cross-app sync features.
- Daily RoundupReplicate adds flux-video-upscale and p-video-avatar, Vercel CLI tools ship, and agent runtimes to test
Replicate released two video models while Vercel pushed CLI and observability updates; industry reports showed Grok issues alongside new routing and agent tools across platforms.