Xiaomi open-sources MiMo-V2.5: 311B A15B 1M-context omnimodal model
Xiaomi has open-sourced MiMo-V2.5, a 311-billion-parameter omnimodal AI model with 15 billion activated parameters and support for up to 1 million tokens of context. The model integrates text, image, video, and audio understanding within a unified architecture and features hybrid attention, multi-token prediction, and efficient FP8 training. It is designed for strong performance in multimodal reasoning, long-context tasks, and agentic workflows.
- ▪MiMo-V2.5 is a sparse Mixture of Experts (MoE) model with 310 billion total parameters and 15 billion activated parameters.
- ▪The model supports a context length of up to 1 million tokens and uses a hybrid attention mechanism to reduce KV-cache storage.
- ▪It includes a 729M-parameter Vision Transformer and a dedicated audio encoder for native multimodal understanding.
- ▪MiMo-V2.5 was trained on approximately 48 trillion tokens using FP8 mixed precision.
- ▪The model incorporates Multi-Token Prediction and agentic reinforcement learning for improved inference and task performance.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | Huggingface |
| Canonical URL | https://huggingface.co/XiaomiMiMo/MiMo-V2.5 |
| Publication time | Tue, 28 Apr 2026 05:25:30 +0000 |
| Retrieval time | 2026-04-28T05:39:00.702Z |
| Last seen | 2026-04-28T05:39:00.702Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | 1eA3AYBROHjp |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
XiaomiMiMo / MiMo-V2.5 like 240 Follow Xiaomi MiMo 3.14k Safetensors English Chinese mimo_v2 multimodal vision-language audio agent video-understanding long-context custom_code Eval Results fp8 License: mit Model card Files Files and versions xet Community 15 MiMo-V2.5 1. Introduction Model Summary 2. Downloads 3. Evaluation Results Multimodal BenchmarksCoding & Agent BenchmarksLong Context Benchmarks4. Model Architecture LLM BackboneVision EncoderAudio Encoder5. Training Process 6.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Huggingface.
Discussion
0 commentsMore from Huggingface
