---
title: "AI models released in October 2025"
type: "release-archive"
canonical_url: "https://theaiengineering.com/releases/2025/10"
markdown_url: "https://theaiengineering.com/releases/2025/10.md"
author: "The AI Engineering Editors"
publisher: "The AI Engineering"
updated: "2025-10-31"
citation: "The AI Engineering Editors. \"AI models released in October 2025.\" The AI Engineering, updated Oct 31, 2025. https://theaiengineering.com/releases/2025/10"
license: "CC BY 4.0"
license_url: "https://creativecommons.org/licenses/by/4.0/"
data_csv: "https://theaiengineering.com/releases.csv"
data_json: "https://theaiengineering.com/releases.json"
---

# AI models released in October 2025

32 releases from 19 labs, 18 with open weights. Dates are the labs' official announcement dates.

## October 2025

### MiniMax Music 2.0

MiniMax Music 2.0 is a music generation model with more lifelike vocals across singing styles and finer instrumental control, generating structurally complete songs with verses, choruses, and bridges of up to five minutes.

- Date: 2025-10-31
- Lab: MiniMax
- Type: Music
- Weights: Proprietary
- Modalities: text → audio
- Available at launch: API, Consumer app
- Permalink: https://theaiengineering.com/releases/2025/10#minimax-music-2-0

Highlights:

- Supports male-female duets and a cappella
- Covers genres including pop, jazz, blues, rock, and folk

Source: [Making Music Creation Accessible to Everyone](https://www.minimax.io/news/minimax-music-20)

### Kimi Linear

Kimi Linear is a hybrid linear attention architecture built on Kimi Delta Attention, a refined version of Gated DeltaNet. Moonshot released 48B base and instruct checkpoints trained on 5.7T tokens.

- Date: 2025-10-30
- Lab: Moonshot AI
- Type: LLM
- Weights: Open weights
- Modalities: text → text
- Context window: 1,000,000 tokens
- Parameters: MoE: 48B total, 3B active
- License: MIT
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#kimi-linear

Highlights:

- Reduces KV cache needs by up to 75%
- Up to 6 times higher decoding throughput at contexts up to 1M tokens
- KDA kernel open-sourced in the flash-linear-attention library

Source: [moonshotai/Kimi-Linear-48B-A3B-Instruct (Hugging Face)](https://huggingface.co/moonshotai/Kimi-Linear-48B-A3B-Instruct)

### MiniMax Speech 2.6

MiniMax Speech 2.6 is a text-to-speech model tuned for voice agents, with end-to-end latency under 250 milliseconds, direct reading of URLs, emails, phone numbers, dates, and amounts, and more natural prosody.

- Date: 2025-10-30
- Lab: MiniMax
- Type: Speech
- Weights: Proprietary
- Modalities: text → audio
- Available at launch: API, Consumer app
- Permalink: https://theaiengineering.com/releases/2025/10#minimax-speech-2-6

Highlights:

- Offered as speech-2.6-hd and speech-2.6-turbo
- Introduces Fluent LoRA for more fluent cloned voices
- Voice cloning across the 40+ languages the model supports

Source: [MiniMax Speech 2.6: The Ultimate Voice Agent Has Arrived](https://www.minimax.io/news/minimax-speech-26)

### gpt-oss-safeguard-120b and gpt-oss-safeguard-20b

gpt-oss-safeguard is a research preview of open-weight safety reasoning models, fine-tuned from gpt-oss, that classify messages, completions, and chats against a developer-provided policy interpreted at inference time.

- Date: 2025-10-29
- Lab: OpenAI
- Type: Reasoning
- Weights: Open weights
- Modalities: text → text
- Parameters: 120b: 117B total, 5.1B active; 20b: 21B total, 3.6B active
- License: Apache-2.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#gpt-oss-safeguard

Highlights:

- Released in two sizes, 120b and 20b, with weights on Hugging Face
- Same Apache 2.0 license as gpt-oss
- Provides chain-of-thought that developers can review to see how decisions were reached
- Developed with ROOST, which launched a model community alongside the release

Source: [Introducing gpt-oss-safeguard | OpenAI](https://openai.com/index/introducing-gpt-oss-safeguard/)

### Amazon Nova Multimodal Embeddings

Amazon Nova Multimodal Embeddings is a single embedding model that maps text, documents, images, video and audio into one semantic space for cross-modal retrieval, agentic RAG and semantic search.

- Date: 2025-10-28
- Lab: Amazon
- Type: Embedding
- Weights: Proprietary
- Modalities: text, image, video, audio → embedding
- Context window: 8,000 tokens
- Available at launch: API
- Permalink: https://theaiengineering.com/releases/2025/10#amazon-nova-multimodal-embeddings

Highlights:

- Supports text, documents, images, video and audio inputs
- Four output dimension sizes trained with Matryoshka Representation Learning
- Up to 8K tokens of context, with text in up to 200 languages

Source: [Amazon Nova Multimodal Embeddings: State-of-the-art embedding model for agentic RAG and semantic search](https://aws.amazon.com/blogs/aws/amazon-nova-multimodal-embeddings-now-available-in-amazon-bedrock/)

### Granite 4.0 Nano

Smallest Granite 4.0 models, released in roughly 350M and 1B sizes as both hybrid H and conventional versions for edge and on-device use.

- Date: 2025-10-28
- Lab: IBM
- Type: Small model
- Weights: Open weights
- Modalities: text → text
- Parameters: 350M to 1.6B
- License: Apache-2.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#granite-4-0-nano

Highlights:

- 1B models support 128K tokens; 350M models support 32K
- Base and instruct checkpoints released
- Released under Apache 2.0

Source: [ibm-granite/granite-4.0-h-1b · Hugging Face](https://huggingface.co/ibm-granite/granite-4.0-h-1b)

### Kaleido-14B-S2V

Kaleido is Z.ai's open-weights multi-subject reference video generation model, which generates video that keeps the identity of several reference subjects consistent. Z.ai released the Kaleido-14B-S2V checkpoints after open-sourcing the code.

- Date: 2025-10-28
- Lab: Z.ai
- Type: Video
- Weights: Open weights
- Modalities: text, image → video
- Parameters: 14B
- License: MIT
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#kaleido-14b-s2v

Highlights:

- Training and inference code released October 22, 2025
- Checkpoints released October 28, 2025
- Weights under the MIT license

Source: [Kaleido (GitHub)](https://github.com/zai-org/Kaleido)

### LFM2-ColBERT-350M

A multilingual late-interaction retriever built on the LFM2 backbone that encodes queries and documents at the token level, supporting cross-lingual search where documents in one language are retrieved with queries in another.

- Date: 2025-10-28
- Lab: Liquid AI
- Type: Embedding
- Weights: Open weights
- Modalities: text → embedding
- Parameters: 353M
- License: LFM Open License v1.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#lfm2-colbert-350m

Highlights:

- Document length 512 tokens and query length 32 tokens per model card
- Positioned as a drop-in replacement in existing RAG pipelines
- Available on Hugging Face with an interactive demo
- Succeeded by LFM2.5-ColBERT-350M in 2026

Source: [LFM2-ColBERT-350M: One Model to Embed Them All](https://www.liquid.ai/blog/lfm2-colbert-350m-one-model-to-embed-them-all)

### MiniMax Hailuo 2.3

MiniMax Hailuo 2.3 is a video generation model building on Hailuo 02, with improved body movement, stylization, and facial micro-expressions, and better response to motion instructions.

- Date: 2025-10-28
- Lab: MiniMax
- Type: Video
- Weights: Proprietary
- Modalities: text, image → video
- Available at launch: API, Consumer app
- Permalink: https://theaiengineering.com/releases/2025/10#minimax-hailuo-2-3

Highlights:

- Launched with Hailuo 2.3 Fast, a faster image-to-video variant
- Broader support for anime, illustration, ink wash, and game CG styles
- Available on the Hailuo AI website and app and the MiniMax API

Source: [MiniMax Hailuo 2.3: A New Level of Complex Video Performance & Media Agent](https://www.minimax.io/news/minimax-hailuo-23)

### Nemotron Nano 12B v2 VL

Multimodal document intelligence model that reads multiple document images such as invoices, receipts and manuals for question answering and extraction.

- Date: 2025-10-28
- Lab: NVIDIA
- Type: Multimodal
- Weights: Open weights
- Modalities: text, image → text
- Context window: 128,000 tokens
- Parameters: 12.6B
- License: NVIDIA Open Model License
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#nemotron-nano-2-vl

Highlights:

- 128K tokens of combined input and output
- Handles multiple document images per request
- Released under the NVIDIA Open Model License

Source: [nvidia/NVIDIA-Nemotron-Nano-12B-v2-VL-BF16 · Hugging Face](https://huggingface.co/nvidia/NVIDIA-Nemotron-Nano-12B-v2-VL-BF16)

### MiniMax M2

MiniMax M2 is an open-weight mixture-of-experts model built for coding and agentic workflows, designed for end-to-end development in tools such as Claude Code and Cursor and for long chains of tool calls.

- Date: 2025-10-27
- Lab: MiniMax
- Type: Coding
- Weights: Open weights
- Modalities: text → text
- Parameters: MoE: 230B total, 10B active
- License: Modified-MIT
- Available at launch: API, Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#minimax-m2

Highlights:

- 230B total parameters with 10B active
- Weights on Hugging Face under a modified MIT license
- Uses interleaved thinking during tool use
- Open-sourced and launched on the MiniMax API the same day

Source: [MiniMax M2 & Agent: Ingenious in Simplicity](https://www.minimax.io/news/minimax-m2)

### LongCat-Video

LongCat-Video is Meituan's 13.6B-parameter foundational video generation model, covering text-to-video, image-to-video, and video continuation in a single model, released with a technical report.

- Date: 2025-10-25
- Lab: Meituan
- Type: Video
- Weights: Open weights
- Modalities: text, image, video → video
- Parameters: 13.6B
- License: MIT
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#longcat-video

Highlights:

- 13.6B parameters
- Single model for text-to-video, image-to-video, and video continuation
- Weights released under MIT

Source: [GitHub - meituan-longcat/LongCat-Video](https://github.com/meituan-longcat/LongCat-Video)

### LTX-2

Unified audio-video foundation model that generates synchronized video and sound in one process, with native 4K at up to 50 fps and clips up to 10 seconds at launch.

- Date: 2025-10-23
- Lab: Lightricks
- Type: Video
- Weights: Open weights
- Modalities: text, image, video, audio → video, audio
- Parameters: 19B
- License: LTX-2 Community License
- Available at launch: Open weights, API, Consumer app
- Permalink: https://theaiengineering.com/releases/2025/10#ltx-2

Highlights:

- Full open weights and code released January 6, 2026
- 19B-parameter checkpoints with FP8 and FP4 variants
- Multi-keyframe conditioning and LoRA fine-tuning

Source: [Release Notes | LTX](https://ltx.io/release-notes)

### Seed3D 1.0

Seed3D 1.0 is a DiT-based 3D generation model that creates high-fidelity, simulation-ready 3D assets with detailed textures and materials from a single image, aimed at world simulators and embodied AI.

- Date: 2025-10-23
- Lab: ByteDance Seed
- Type: Image
- Weights: Proprietary
- Modalities: image → 3d
- Available at launch: API
- Permalink: https://theaiengineering.com/releases/2025/10#seed3d-1-0

Highlights:

- Geometry generation model has 1.5B parameters
- Volcano Engine describes outputs as meshes with PBR materials
- API released alongside the technical report
- Extensible to scene generation

Source: [Seed3D 1.0 Released: Generate High-Fidelity 3D Models from Single Images, Featuring SOTA Texturing](https://seed.bytedance.com/en/blog/seed3d-1-0-released-generate-high-fidelity-3d-models-from-single-images-featuring-sota-texturing)

### HunyuanWorld-Mirror (HunyuanWorld 1.1)

HunyuanWorld-Mirror, released as HunyuanWorld 1.1, is an open feed-forward model for 3D geometric prediction that creates 3D worlds from videos or multi-view images, optionally using camera pose, intrinsics, and depth priors.

- Date: 2025-10-22
- Lab: Tencent
- Type: Multimodal
- Weights: Open weights
- Modalities: image, video → 3d
- License: Tencent HunyuanWorld-Mirror Community License Agreement
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#hunyuanworld-mirror

Highlights:

- Outputs point clouds, multi-view depths, camera parameters, surface normals, and 3D Gaussians
- Training and evaluation code followed on November 7, 2025

Source: [GitHub - Tencent-Hunyuan/HunyuanWorld-Mirror: \[ICML 2026\] WorldMirror: Fast and Universal 3D reconstruction model for versatile tasks](https://github.com/Tencent-Hunyuan/HunyuanWorld-Mirror)

### LFM2-VL-3B

A 3B-parameter vision-language model built on the LFM2-2.6B backbone with a SigLIP2 400M NaFlex encoder, adding more capacity to the LFM2-VL family and extending visual understanding to ten languages.

- Date: 2025-10-22
- Lab: Liquid AI
- Type: Multimodal
- Weights: Open weights
- Modalities: text, image → text
- Context window: 32,768 tokens
- Parameters: 3B
- License: LFM Open License v1.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#lfm2-vl-3b

Highlights:

- Processes images at native resolution with variable aspect ratios
- Adjustable vision tokens per image to balance speed and quality
- Available on Hugging Face under the LFM Open License and on LEAP

Source: [LFM2-VL-3B: A New Efficient Vision-Language for the Edge](https://www.liquid.ai/blog/lfm2-vl-3b-a-new-efficient-vision-language-for-the-edge)

### olmOCR 2

Updated OCR model for English digitized print documents, built on Qwen2.5-VL-7B and trained with unit test rewards for document correctness.

- Date: 2025-10-22
- Lab: Ai2
- Type: Multimodal
- Weights: Open weights
- Modalities: text, image → text
- Parameters: 7B
- License: Apache-2.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#olmocr-2

Highlights:

- Released as olmOCR-2-7B-1025
- Fine-tuned on olmOCR-mix-1025 with 270,000 PDF pages
- Released under Apache 2.0

Source: [olmOCR 2: Unit test rewards for document OCR | Ai2](https://allenai.org/blog/olmocr-2)

### DeepSeek-OCR

DeepSeek-OCR is an open-weights vision-language model that explores contexts optical compression, studying the role of vision encoders from an LLM-centric viewpoint. It converts document images to text and markdown.

- Date: 2025-10-20
- Lab: DeepSeek
- Type: Multimodal
- Weights: Open weights
- Modalities: text, image → text
- License: MIT
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#deepseek-ocr

Highlights:

- Released on GitHub and Hugging Face on October 20, 2025
- Supported in upstream vLLM from October 23, 2025
- Weights under the MIT License

Source: [DeepSeek-OCR: Contexts Optical Compression (GitHub)](https://github.com/deepseek-ai/DeepSeek-OCR)

### PaddleOCR-VL

PaddleOCR-VL is Baidu's open document-parsing system built around PaddleOCR-VL-0.9B, a compact vision-language model that pairs a NaViT-style dynamic-resolution visual encoder with the ERNIE-4.5-0.3B language model.

- Date: 2025-10-16
- Lab: Baidu
- Type: Multimodal
- Weights: Open weights
- Modalities: image, text → text
- Parameters: 0.9B
- License: Apache-2.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#paddleocr-vl

Highlights:

- Supports 109 languages
- Recognizes text, tables, formulas, and charts, including handwriting and historical documents
- Uses PP-DocLayoutV2 for layout analysis and reading order, with Markdown and JSON output

Source: [PaddleOCR-VL: Boosting Multilingual Document Parsing via a 0.9B Ultra-Compact Vision-Language Model](https://ernie.baidu.com/research/posts/paddleocr-vl/)

### Claude Haiku 4.5

Claude Haiku 4.5 is Anthropic's small model, which Anthropic says matches Claude Sonnet 4 on coding at lower cost and higher speed, aimed at real-time, low-latency work and multi-agent setups that run subtasks in parallel.

- Date: 2025-10-15
- Lab: Anthropic
- Type: Small model
- Weights: Proprietary
- Modalities: text, image → text
- Context window: 200,000 tokens
- Available at launch: API, Consumer app, Cloud platforms
- Permalink: https://theaiengineering.com/releases/2025/10#claude-haiku-4-5

Highlights:

- API model ID claude-haiku-4-5 (snapshot claude-haiku-4-5-20251001)
- 200K-token context window and 64K max output tokens
- Available in the Claude apps, Claude Code, the Claude API, Amazon Bedrock, and Google Cloud Vertex AI

Source: [Introducing Claude Haiku 4.5](https://www.anthropic.com/news/claude-haiku-4-5)

### Doubao-Seed-1.6-lite

Volcano Engine upgraded Doubao-Seed-1.6 with four selectable thinking lengths (Minimal, Low, Medium, High) and released Doubao-Seed-1.6-lite, a smaller and faster deep-thinking model for enterprise workloads.

- Date: 2025-10-15
- Lab: ByteDance Seed
- Type: Reasoning
- Weights: Proprietary
- Modalities: text → text
- Available at launch: API
- Permalink: https://theaiengineering.com/releases/2025/10#doubao-seed-1-6-lite

Highlights:

- Adds Minimal, Low, Medium, and High thinking-length settings
- Volcano Engine calls it the first model in China with native graded thinking-length control
- Doubao-Seed-1.6-lite released the same day

Source: [豆包大模型1.6升级：国内首个原生支持“分档调节思考长度”的大模型-发布中心-火山引擎](https://www.volcengine.com/news/detail/23)

### Doubao-Seed-1.6-Rerank

Doubao-Seed-1.6-Rerank is an all-modality reranking model that reorders retrieved text, image, and video results, integrated into Volcano Engine's VikingDB vector database alongside Doubao-Seed-1.6-Embedding for end-to-end retrieval.

- Date: 2025-10-15
- Lab: ByteDance Seed
- Type: Reranker
- Weights: Proprietary
- Modalities: text, image, video → text
- Available at launch: API
- Permalink: https://theaiengineering.com/releases/2025/10#doubao-seed-1-6-rerank

Highlights:

- Reranks mixed text, image, and video candidates
- Understands negation and comparison logic in queries
- Produces stable relevance scores suited to threshold filtering

Source: [向量库支持全模态Doubao-Seed-1.6-Rerank模型，实现VectorDB+Embedding+Rerank全环节融合-发布中心-火山引擎](https://www.volcengine.com/news/detail/20)

### Seed-TTS 2.0 and Seed-ICL 2.0

Seed-TTS 2.0 and Seed-ICL 2.0 are Doubao's speech synthesis and voice cloning models, introducing a query-response conversational synthesis approach that uses dialogue context and natural-language instructions to control tone, emotion, and speaking rate.

- Date: 2025-10-15
- Lab: ByteDance Seed
- Type: Speech
- Weights: Proprietary
- Modalities: text → audio
- Available at launch: API
- Permalink: https://theaiengineering.com/releases/2025/10#seed-tts-2-0

Highlights:

- Takes the user's query and background information as context for delivery
- Speaking rate, emotion, and style can be steered with text instructions
- Available in the Doubao Speech console on Volcano Engine

Source: [从文本朗读到情感演绎，豆包语音模型重磅升级-发布中心-火山引擎](https://www.volcengine.com/news/detail/14)

### Veo 3.1

Veo 3.1 is Google DeepMind's upgrade to Veo 3, bringing richer generated audio, stronger prompt adherence for image-to-video and more narrative control. It launched with a Veo 3.1 Fast variant across Google's apps and APIs.

- Date: 2025-10-15
- Lab: Google
- Type: Video
- Weights: Proprietary
- Modalities: text, image → video, audio
- Available at launch: Consumer app, API, Cloud platforms
- Permalink: https://theaiengineering.com/releases/2025/10#veo-3-1

Highlights:

- Flow features Ingredients to Video, Frames to Video and Extend now include audio
- Veo 3.1 and Veo 3.1 Fast entered public preview in the Gemini API on launch day
- Generally available on Vertex AI from November 17, 2025
- Also available in the Gemini app

Source: [Introducing Veo 3.1 and advanced capabilities in Flow](https://blog.google/technology/ai/veo-updates-flow/)

### MAI-Image-1

MAI-Image-1 is Microsoft AI's first image generation model developed entirely in-house. It later became available in Bing Image Creator and in Copilot Audio Expressions Story Mode.

- Date: 2025-10-13
- Lab: Microsoft
- Type: Image
- Weights: Proprietary
- Modalities: text → image
- Available at launch: Consumer app
- Permalink: https://theaiengineering.com/releases/2025/10#mai-image-1

Highlights:

- Microsoft's first in-house text-to-image model
- Added to Bing Image Creator and Copilot on November 4, 2025
- Available in countries that can access Bing Image Creator and Copilot Labs

Source: [Introducing MAI-Image-1, debuting in the top 10 on LMArena](https://microsoft.ai/news/introducing-mai-image-1-debuting-in-the-top-10-on-lmarena/)

### Jamba Reasoning 3B

Compact open-source reasoning model built on AI21's hybrid SSM-Transformer architecture, designed to run on phones and computers for long-context reasoning.

- Date: 2025-10-08
- Lab: AI21 Labs
- Type: Reasoning
- Weights: Open weights
- Modalities: text → text
- Context window: 256,000 tokens
- Parameters: 3B
- License: Apache-2.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#jamba-reasoning-3b

Highlights:

- Handles 256K tokens and can stretch to 1M
- Runs locally via llama.cpp or LM Studio
- Available on Hugging Face and Kaggle

Source: [Introducing Jamba Reasoning 3B: Tiny Model, Huge Possibilities](https://www.ai21.com/blog/introducing-jamba-reasoning-3b/)

### Gemini 2.5 Computer Use

The Gemini 2.5 Computer Use model is a specialized model built on Gemini 2.5 Pro's visual understanding and reasoning that powers agents able to interact with user interfaces in browsers and on mobile.

- Date: 2025-10-07
- Lab: Google
- Type: Multimodal
- Weights: Proprietary
- Modalities: image, text → text
- Context window: 128,000 tokens
- Available at launch: API, Cloud platforms
- Permalink: https://theaiengineering.com/releases/2025/10#gemini-2-5-computer-use

Highlights:

- Public preview in the Gemini API on Google AI Studio and Vertex AI
- Model page lists a 128,000 input token limit and 64,000 output tokens
- Computer Use later became a tool in Gemini 3 and 3.5 models, reaching general availability on Oct 30, 2026

Source: [Introducing the Gemini 2.5 Computer Use model](https://blog.google/innovation-and-ai/models-and-research/google-deepmind/gemini-computer-use-model/)

### LFM2-8B-A1B

Liquid AI's first on-device mixture-of-experts model, an early checkpoint that adds sparse MoE feed-forward layers to the LFM2 backbone, with 8.3B total and 1.5B active parameters per token.

- Date: 2025-10-07
- Lab: Liquid AI
- Type: Small model
- Weights: Open weights
- Modalities: text → text
- Context window: 32,768 tokens
- Parameters: MoE: 8.3B total, 1.5B active
- License: LFM Open License v1.0
- Available at launch: Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#lfm2-8b-a1b

Highlights:

- 32 experts per MoE block with top-4 routing; 18 convolution and 6 GQA blocks
- Pre-trained on about 12T tokens with more code than LFM2-2.6B
- Weights and GGUF quants on Hugging Face; testable on Liquid Playground
- Later replaced by LFM2.5-8B-A1B (May 2026)

Source: [LFM2-8B-A1B: An Efficient On-device Mixture-of-Experts](https://www.liquid.ai/blog/lfm2-8b-a1b-an-efficient-on-device-mixture-of-experts)

### GPT Image 1 Mini

GPT Image 1 Mini is a cost-efficient version of GPT Image 1 for image generation and editing, released in the API at OpenAI DevDay 2025 alongside new audio and video models.

- Date: 2025-10-06
- Lab: OpenAI
- Type: Image
- Weights: Proprietary
- Modalities: text, image → image
- Available at launch: API
- Permalink: https://theaiengineering.com/releases/2025/10#gpt-image-1-mini

Highlights:

- API model ID gpt-image-1-mini on the image generation and edit endpoints
- Released at OpenAI DevDay on October 6, 2025
- Accepts text and image input and outputs images, per the API model page

Source: [Changelog | OpenAI API](https://developers.openai.com/api/docs/changelog)

### gpt-realtime-mini and gpt-audio-mini

gpt-realtime-mini and gpt-audio-mini are cost-efficient versions of gpt-realtime and gpt-audio for speech-to-speech applications, released at OpenAI DevDay 2025 for the Realtime API and Chat Completions API respectively.

- Date: 2025-10-06
- Lab: OpenAI
- Type: Speech
- Weights: Proprietary
- Modalities: text, audio, image → text, audio
- Available at launch: API
- Permalink: https://theaiengineering.com/releases/2025/10#gpt-realtime-mini

Highlights:

- gpt-realtime-mini runs in the Realtime API; gpt-audio-mini in Chat Completions
- Context: 32,000 tokens for gpt-realtime-mini and 128,000 for gpt-audio-mini
- Image input is supported by gpt-realtime-mini only, per the API model pages
- Updated snapshots dated 2025-12-15 followed in December 2025

Source: [Changelog | OpenAI API](https://developers.openai.com/api/docs/changelog)

### Granite 4.0

Fourth-generation Granite language models that add hybrid H variants, including mixture-of-experts models, alongside a conventional dense Micro model, all with 128K-token context.

- Date: 2025-10-02
- Lab: IBM
- Type: LLM
- Weights: Open weights
- Modalities: text → text
- Context window: 128,000 tokens
- Parameters: H-Small 32B (9B active), H-Tiny 7B (1B active), H-Micro 3B, Micro 3B
- License: Apache-2.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#granite-4-0

Highlights:

- H-Small activates 9B of 32B parameters
- Base and instruct checkpoints released
- Released under Apache 2.0

Source: [ibm-granite/granite-4.0-h-small · Hugging Face](https://huggingface.co/ibm-granite/granite-4.0-h-small)

### LFM2-Audio-1.5B

An end-to-end audio and text foundation model that extends LFM2-1.2B to accept and generate both speech and text, covering conversational chat, speech recognition, text-to-speech, and audio classification in one 1.5B-parameter model.

- Date: 2025-10-01
- Lab: Liquid AI
- Type: Speech
- Weights: Open weights
- Modalities: text, audio → text, audio
- Context window: 32,768 tokens
- Parameters: 1.5B
- License: LFM Open License v1.0
- Available at launch: Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#lfm2-audio-1-5b

Highlights:

- Continuous embeddings for audio input; discrete audio tokens for output
- Tokenizer-free input: raw waveform chunked into segments of about 80 ms
- Supported language per model card: English
- Available on Hugging Face and Liquid Playground

Source: [LFM2-Audio: An End-to-End Audio Foundation Model](https://www.liquid.ai/blog/lfm2-audio-an-end-to-end-audio-foundation-model)

## Cite and reuse this data

These release records are free to reuse under CC BY 4.0, including commercially. Credit The AI Engineering and link to this page's URL. Dates are the labs' official announcement dates, and every record links to its source. List fields in the CSV are separated by " | ".

The AI Engineering Editors. "AI models released in October 2025." The AI Engineering, updated Oct 31, 2025. https://theaiengineering.com/releases/2025/10

HTML link:

```html
<a href="https://theaiengineering.com/releases/2025/10">AI models released in October 2025</a> (The AI Engineering)
```

- Title: AI models released in October 2025
- Author: The AI Engineering Editors
- Publisher: The AI Engineering
- Updated: 2025-10-31
- URL: https://theaiengineering.com/releases/2025/10
- Markdown: https://theaiengineering.com/releases/2025/10.md
- License: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/)
- Full dataset as CSV: https://theaiengineering.com/releases.csv
- Full dataset as JSON: https://theaiengineering.com/releases.json
