These models hold a spoken conversation: they listen and reply with speech without a separate recognition, text and synthesis pipeline. They power phone voice bots and in-app assistants. Check response latency, support for your languages, the license and GPU requirements.
NVIDIA models for agents and reasoning, optimized to run fast on its GPUs. Nemotron 3 is a Mamba and MoE hybrid from 4B to 550B; Nano Omni handles video, audio and images (English only). The new Nemotron 3.5 Super VL (120B, 12B active, up to 1M-token context) reads text, images and video up to two minutes, with a focus on documents and agents; the compressed Bolt (74B, 9B active) fits into 24 GB of video memory. Super VL languages: English, Spanish, French, German, Italian, Japanese. Dory 1.0 (4B, up to 256K-token context, English) runs the same block of layers several times, and the number of passes can be changed per task: more passes give more accuracy without a new model and without long reasoning in the text.
Small decision models: from text, an image or a voice clip they answer yes or no, pick an option or give a rating in one pass, without writing text. d1-3B sees images, d1-omni-600M also hears speech up to 30 seconds. Russian is listed. The open d1-2B (text, pictures and video, based on LFM3-VL-2B) and d1-Omni-2B (sound as well) decide in milliseconds, 5.7 ms on an H100 by the authors measurements; besides choosing an answer they pull out text, numbers and dates with their positions and find objects in a picture. It is not a chat model.
routing requests and tickets to teams
checking a photo for defects or violations against a list of questions
Models with a new architecture for on-device use: fast on a regular CPU and on phones. Versions for data extraction, RAG and tools, plus LFM2.5-VL for images and voice LFM2.5-Audio. The largest, LFM2.5-24B-A2B (2.45B active, 131K-token context, Russian in the language list), always reasons before answering and targets fast on-device agents. In October the LFM3 generation came out on a new tri-hybrid architecture: LFM3-1.7B for local agents (256 thousand tokens of context, about 1 GB of memory at 4 bits, pre-trained on 22 trillion tokens), LFM3-VL-2B for images, LFM3-Audio-2B for speech and LFM3-Omni-2B, which understands text, images, video and sound and answers in text and speech. The authors do not recommend it for complex code and multi-step reasoning.
Meta's largest open model: 816B parameters, 42B active per token, a 1 million token context. It understands text, images, video and audio and answers in text. Built for programming and long agentic tasks. Trained on data in more than 100 languages. Access to the weights on Hugging Face is granted manually.
an agent for working with code and long tasks with tool calling
analyzing screenshots, charts, documents, video and recordings
One compact model turns a document page, chart, flowchart, geometry figure, audio or video into structured JSON: with block coordinates, tables, formulas, speech transcription and timestamps.
parsing scans and PDFs with tables and formulas
turning charts and flowcharts into a table or diagram
Decision models: they do not write text but read a situation (text, JSON, an image, video frames, and for Clef-Omni also audio) and return in one pass the probability of each answer to typed questions: yes or no, pick one, rate on a scale. The agent gets ready numbers instead of parsing free text.
sorting requests and tickets into categories with a probability
checking a document or photo against a list of questions (is everything stated, is there a violation)
choosing the next step of an agent without parsing a text answer
StepFun MoE models built for fast, low-cost work: with 196 billion parameters, Step-3.5/3.7-Flash use about 11 billion per token. Compact Step3-VL-10B for images and voice Step-Audio 2 mini are available. The flagship Step 5 Preview (600B, 27B active, up to 1M-token context, text and images as input) is built for agents, programming and finance tasks.
High-load agents
Analysis of documents with diagrams and screenshots
Sber open models with strong Russian language support and local context, from 10B-A1.8B to 702B, all MIT. GigaChat3.1-Audio handles recordings up to two hours; GFusion is a fast diffusion text version.
Russian-language employee assistant on your own server
Xiaomi models for reasoning and agents: from the compact MiMo-7B to MiMo-V2.6-Pro with 1.02 trillion parameters. The larger versions understand text, images, video and audio, with a 1M token context. Languages: English and Chinese.
Tiny audio-understanding models for smartphones: they listen to speech, music and ambient sounds and answer in text - describing a recording and answering questions about it. They run on the device itself; prompts and answers are in English - no other languages are present in the training data.
Models from Meituan, China's largest delivery service. LongCat-Flash adjusts compute to query complexity; LongCat-2.0 has 1.6 trillion parameters under MIT. Omni models (Flash-Omni, Next) and AudioDiT speech synthesis too.
Models that listen to speech, sounds and music and answer questions about them. Audio Flamingo Next handles recordings up to 30 minutes. Research use only.
A voice assistant that listens and speaks at the same time, with no delay for recognition and synthesis. Hibiki does simultaneous speech-to-speech translation between several European languages.
A voice conversation partner based on Moshi that listens and speaks at the same time and can be interrupted. The role is set by text, the voice by a sample recording. English only.
Russian-language fine-tunes of open models (Mistral, Qwen, Llama) by the independent Vikhr team, with compact versions for a regular PC. Borealis is an audio model for recognizing and understanding Russian speech.
Russian-language assistant on your own PC or server
A general-purpose audio model: speech recognition, answering questions about sounds, detecting emotions and voice dialogue. Trained on 13 million hours of audio; languages are English and Chinese.
Speech and text translation across roughly a hundred languages, including Russian: speech to text, speech to speech, and streaming translation that keeps intonation.
