multimodal

Salesforce introduced Koa Decisions on November 6, 2026. The decision model is built from Google's open-weight Gemma and accepts text, images and audio. It is the next version of HyperClassifier.

Sage Sci Perception (书生·元征) connects scientific encoders to a language model through trainable projectors. Published results include 97.35% accuracy on BEACON ncRNA RNA family classification and 49.75% Acc@1 on QM9S joint-spectrum structure identification.

Ant Group released Ling-3.2-flash on November 5, a native multimodal model that accepts text, images, and video. It has 560 billion total parameters, activates 25 billion per inference step, and supports a one-million-token context window.

Google published a Gemini walkthrough on November 5, 2026, showing how a photograph of a paper sketch can become an interactive app in Canvas. The demonstration covers uploading the image, generating code, testing the result and refining it through conversation.

Google released Gemini Omni 1.2 Flash on November 5, 2026, a native multimodal video generation and editing model that succeeds Gemini Omni 1.1 Flash. It generates clips up to 15 seconds with native audio at resolutions from 360p to 4K. OpenRouter lists it at $1.50 input and $9 output per 1M with a 131K context window.

Google released Gemini Omni 1.1 Flash on Nov 4, 2026, a native multimodal video generation and editing model that is the generally available successor to Gemini Omni Flash Preview. It outputs clips from 360p drafts up to 4K with native audio, and is listed on OpenRouter at $1.50 input / $9 output per 1M tokens with a 131K context window.

SenseTime has released SenseNova 6.8 Flash (日日新), a multimodal agent model built around "delegated intelligence": the user sets a goal, the AI runs the full workflow, and the human only reviews. According to SenseTime's results, it ranks first on TeamBench and DataAnalysis2.0 and is free for a limited time.

Z.ai released GLM-5.3-FlashX-Legal (GLM-Legal) on November 2, 2026, a legal-domain variant of GLM-5.3-FlashX with a 1M-token context window. It is priced at $0.37 per 1M input tokens and $1.25 per 1M output tokens on OpenRouter.

Meituan's LongCat team released LongCat-2.5 on November 2, a native multimodal model with 1.6 trillion total parameters and about 48 billion active per token. Meituan claims 88.4 on Terminal Bench v2.1, up from 70.8 for LongCat-2.0. The figures are unverified by third parties.

Nace AI has open-sourced Drex Max v1, a 27B multimodal decision model. MarkTechPost reports it scores 67.52 and runs locally on Apple Silicon.

SenseTime open-sourced SenseNova-Loop-9B-Preview on October 30 under Apache 2.0. The multimodal model stores 32 decoder layers but runs 56 layer executions per forward pass, and SenseTime claims it matches models of around 27 billion parameters.

Articul8 has introduced Bivia, a decision-making model that lets enterprise AI agents make structured decisions without calling a generative model at every step. It is available by API as a standalone product or inside Articul8's enterprise AI platform. Articul8 reports roughly 70% workflow completion in an internal evaluation, against roughly 30% for the alternative tested.

Tencent plans to release Hy4 and Hy4 flash soon, with both models supporting visual understanding, according to senior executive vice president Dowson Tong. He spoke at the Tencent Global Digital Ecosystem Conference on October 29.

Liquid AI has released LFM3-Omni-2B, its first omni-modal model in the LFM3 family, with 2.17B total parameters and a 32K-token context. It accepts text, images, video and audio, and outputs text, 24 kHz speech, or both interleaved.

Liquid AI has released LFM3, a family of six 2B-class models built from LFM3-1.7B-Base, covering text, vision, audio, omni and decision tasks. According to Liquid AI, LFM3-1.7B outperforms LFM2.5-2.6B with 35% fewer parameters and offers a 256K-token context window.

vLLM Semantic Router has released d3-ultra, a 27B open-weight multimodal decision model. It reads text, images and video once and returns probabilities for every answer. The team reports a score of 87.2 on its SR-Text test.

iFLYTEK Healthcare announced on October 22 a full upgrade of its Spark medical large model to V3.7, and reports a score of 99.6 on MedBench 5.0. The company says the model now combines speech, medical imaging and language understanding for clinical agent use. The benchmark and usage figures are company-reported.

MiniMax has published MiniMax-M3.1-Flash on Hugging Face: a native multimodal mixture-of-experts model with about 428B total parameters, 23B activated per token, and a context window of up to 1M tokens. It builds on MiniMax-M3 with sparse attention, quantization-aware training and DSpark speculative decoding.

Xiaomi discontinued MiMo-V2.5-Pro and MiMo-V2.5 at 10 a.m. Beijing time on October 21. Their API services now run on the MiMo-V2.6 family, and requests to the old model names have been routed to MiMo-V2.6-Pro and MiMo-V2.6-Flash since October 14.

Xiaomi's SeerRay-Lab has open-sourced Xiaomi-Lens-0, a 4B multimodal model built on Qwen3.5-4B. The model emits mask tokens for pixel-level image regions in the middle of an answer. The lab also released a benchmark called LensBench.

NVIDIA has released Nemotron 3.5 Embed (8B BF16, 1B BF16, 1B NVFP4), Nemotron 3.5 Rerank 8B BF16 and Nemotron 3.5 Parse (631M parameters) as open, commercially available models. NVIDIA reports that the 8B embedder ranks first among dense retrievers on ViDoRe v3.1 with 61.17 nDCG@10, and that adding the reranker cuts agent token cost per query by roughly 13-14%.

DeepSeek has released V4.1 Flash 1017, an October 17, 2026 snapshot built on its new Causal Encoder-Decoder architecture. It uses a 552B-parameter sparse MoE backbone with a 1M-token context window, native image understanding, and pricing of $0.15 input / $0.60 output per 1M tokens on OpenRouter.

Elorian, the Palo Alto startup that raised a $55M seed at a $300M valuation, has released Glimpse as a research preview with API access at $0.60 per 1M input tokens and $3 per 1M output tokens. The model can crop, pull video frames, and run code mid-reasoning. Benchmark claims come from Elorian's own evaluations.

Moondream 3 Preview is a vision-language model with 9B total and 2B active parameters (mixture-of-experts), a 32K-token context window and an optional reasoning mode. OpenRouter lists it at $0.30 per 1M input tokens and $2.50 per 1M output tokens, served by Prime Intellect.

Moondream has released Moondream 3.1, a mixture-of-experts vision-language model with 9B total and 2B active parameters and a 32K-token context window. It is listed on OpenRouter at $0.30 per 1M input tokens and $2.50 per 1M output tokens, with an optional reasoning mode.

Nvidia has released Nemotron 3.5 Super VL, an open multimodal Mixture-of-Experts model with 120 billion total and 12 billion active parameters, built for continuously running enterprise agents. Two compact Bolt models for local PCs and edge devices launch alongside it, though the 4 GB variant ships later this month.

Unbiased has released Pareto 26.9, a multimodal composite model with a 262K-token context window, priced at $2.50 per 1M input tokens and $7.50 per 1M output tokens. It is listed on OpenRouter with a Oct 13, 2026 release date. A cheaper 1.0M-context preview, Pareto 26.10, is already listed alongside it.

OpenBMB has released MiniCPM-V 4.7 1B, a 1B-parameter multimodal model for single-image, multi-image and video understanding on phones, tablets and local hardware. OpenBMB claims it substantially outperforms Qwen3.5 0.8B in overall score and inference efficiency in Instruct mode. Benchmark figures, context window and training cutoff are not given in the available model card text.

OpenBMB has released MiniCPM-V 4.7 35B-A3B, the first mixture-of-experts model in the MiniCPM-V series, with open weights on Hugging Face. The company claims it runs at 13+ tokens/s using 3 GB of memory on an iPhone 17 Pro Max, and that it beats Qwen3.6 35B-A3B on overall score in less inference time in Instruct mode.

Liquid AI released two open-weight decision models, d1-3B (text and image) and the experimental d1-omni-600M (text with image or audio). Unlike generative models, they answer in a single forward pass, and Liquid AI claims d1-3B scores 48.57 on its Decision Index 0.2.1, ahead of all 4B and 9B models it tested.

Unsloth has released GGUF quantizations of EmbeddingGemma 2, Google DeepMind's open 740M-parameter embedding model for text, images, video and audio. The base model maps all four modalities into one 768-dimensional space with an 8K context window under Apache 2.0.

Google announced EmbeddingGemma 2, a 740M-parameter natively multimodal embedding model built on the Gemma 4 architecture and released under Apache 2.0. Google says the quantized model needs about 191MB of active RAM for text-only weights and about 567MB for the full multimodal model on a Pixel 11 Pro. Google also launched a Mac app, AI Edge Foresight, to demonstrate it.

Google's Nano Banana 2.1 (Gemini Nano Banana 2.1) is an image generation and editing model on the Flash tier, listed on OpenRouter at $1.50 input and $30 output per 1M tokens with a 66K context window. It supports 1K, 2K, and 4K output and succeeds Nano Banana 2 and Nano Banana Pro, according to the listing.

Mistral AI has launched a public preview of Mistral Large 4, a 1-trillion-parameter natively multimodal model with 49 billion active parameters. The preview API is live on Mistral Studio, and open weights are promised by the end of October 2026. Pricing and context window have not been disclosed.

Reka AI has released a research preview of Rho-1, a 19-billion-parameter omni-model that processes and generates text, images, video, and robot control actions in a single network. Reka says it uses no tool calls or external models. Context window, pricing, and benchmark scores have not been disclosed.

Google began rolling out Guided Vision in Gemini Live on compatible Android devices on October 1, 2026. The feature uses the phone camera to give real-time audio descriptions of surroundings and objects, with follow-up questions supported. Google warns it should not be used for navigation or obstacle detection.

Cloudflare published Clef on Hugging Face: a 27B multimodal model that takes a state and a schema of typed questions and returns a probability for every allowed option in a single forward pass. It is post-trained from Qwen3.8-27B and released under Apache-2.0. Benchmark results are from Cloudflare's internal Decision Index 0.2.1 run.

Unbiased has listed Pareto 26.10 Preview on OpenRouter, a multimodal composite model with a 1.0M-token context window priced at $0.80 input and $3.20 output per 1M tokens. The company says it targets research, coding, and agentic workflows, and warns the preview may change without notice. No benchmark scores have been published.

Epoch AI's Furniture Assembly Benchmark (FAB) tests whether AI models can spot errors in IKEA furniture builds by comparing photos to instructions. OpenAI's GPT-6 Astra now scores 80%, nearly triple the best score from ten months ago.

Meta Superintelligence Labs has released Muse Glimmer 30B, a dense open-weight model distilled from its larger Muse Spark system and tuned for agentic workflows on consumer hardware. The model supports 131K context, image understanding, and over 100 languages at $0.30/$1.10 per 1M input/output tokens.

Xiaomi's MiMo team has released MiMo-V2.6-Pro-RL, a 1.02-trillion-parameter sparse mixture-of-experts model with 42B active parameters, 1M-token context, and native text/image/video/audio processing. The model was trained via a single mixed reinforcement learning run spanning coding, agentic, visual, and cybersecurity tasks, with benchmark scores that Xiaomi claims approach or match Claude Opus 5 and GPT-5.6 on several agentic and coding tests.

Xiaomi's MiMo-V2.6-Pro-UltraSpeed is a fast-inference edition of the company's 1T-parameter flagship MiMo-V2.6-Pro, delivering roughly 10x the output speed at matching quality. It retains the 1M-token context window and native multimodal capabilities, priced at $4.35/$8.70 per 1M input/output tokens.

Xiaomi has released MiMo-V2.6-Flash, an open-source Mixture-of-Experts model with 309B total parameters and 15B activated per token, featuring a 1M-token context window and native multimodal capabilities. Priced at $0.14 per 1M input tokens and $0.28 per 1M output tokens, it targets agentic coding and long-horizon task workflows.

Xiaomi has released MiMo-V2.6-Pro, a flagship foundation model exceeding 1 trillion parameters with a 1M-token context window and native multimodal support. The model is priced at $0.435 per 1M input tokens and $0.87 per 1M output tokens, targeting agentic and long-horizon tasks.

Z.ai has released GLM-5.3-FlashX, a high-speed variant of GLM-5.3-Flash built on a hybrid sparse and linear attention architecture with 320B total parameters (18B active). The model supports a 1M-token context window and claims inference speeds of up to 200 tokens per second.

Google DeepMind has released Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking, two voice-dialogue models that reason and execute background tasks without interrupting conversation. Google claims the Extended Thinking model ranks #1 on Artificial Analysis' Speech to Speech Quality Index with a score of 82.6.

OpenRouter is listing a model called 'OpenAI GPT Astra Latest' with a 1.05 million token context window and $10/$50 per-million-token pricing. OpenAI has made no public announcement, and the listing's own description says it is an auto-redirecting alias rather than a fixed model.

OpenRouter has added a listing called '~openai/gpt-sol-latest,' described as an alias that always points to the newest model in an undisclosed 'GPT Sol' family from OpenAI. The listing shows a 1050K token context window and pricing of $2.00 per million input tokens and $10.00 per million output tokens, but OpenAI has not publicly confirmed a model line by this name.

Alibaba's Qwen team released Qwen3.8 Max (0902), a 2.4-trillion-parameter mixture-of-experts model with a 1M-token context window that accepts text, image, and video input. The snapshot is post-trained for coding, agentic workflows, and long-horizon task execution, priced at $2/$6 per 1M input/output tokens.

Meta has released Muse Spark 1.3 Contributor, described as the cost-efficient contributor tier of its multimodal reasoning model line. The model offers a 1 million token context window at $0.10 per 1M input tokens and $0.20 per 1M output tokens, targeting experimentation and early-stage agentic workflows.