multilingual

Mistral AI has released Voxtral TTS, a 4B parameter text-to-speech model supporting 9 languages including English, French, German, Spanish, Dutch, Portuguese, Italian, Hindi, and Arabic. The model achieves 70ms latency for typical inputs and can clone voices from as little as 3 seconds of audio, priced at $0.016 per 1,000 characters.

Mistral AI announced a partnership with Agence France-Presse to integrate AFP's newswire content into Le Chat. The assistant will access 2,300 daily stories across six languages—French, English, Spanish, Portuguese, German, and Arabic—from AFP's network of 1,700 journalists.

ServiceNow released AU-Harness, the first comprehensive benchmark for code-switched speech recognition in enterprise voice agents, testing seven ASR systems including ElevenLabs, Gemini, and AssemblyAI. The benchmark covers 918 utterances across Spanish-English, French-English, Canadian French-English, and German-English, measuring Word Error Rate (WER), Semantic WER (SWER), and Answer Error Rate (AER). ElevenLabs Scribe V2 achieved the lowest WER across all language pairs, followed closely by AssemblyAI Universal-3 Pro.

NVIDIA has released Nemotron 3.5 Content Safety, a 4B-parameter model built on Google Gemma 3 4B IT that provides multimodal safety classification across approximately 140 languages. The model includes a 128K context window, custom enterprise policy enforcement, auditable reasoning traces, and is releasing its training dataset.

NVIDIA released Nemotron 3.5 ASR, a 600M-parameter speech-to-text model supporting 40 language-locales from a single checkpoint. The model achieves 0.07 seconds to final transcript after speech ends and ranks 2nd in latency among streaming ASR models according to Artificial Analysis benchmarks.

Mistral AI has released Magistral, its first reasoning model line, in two versions: Magistral Small (24B parameters, Apache 2.0) and Magistral Medium (enterprise). Magistral Medium scored 73.6% on AIME2024 (90% with majority voting at 64 samples), while the open-source Small version achieved 70.7% (83.3% with voting).

Mistral AI has released Saba, a 24B-parameter model trained specifically for Arabic and South Asian languages including Tamil. The model runs on single-GPU systems at over 150 tokens per second and is available via API or for on-premises deployment.

Mistral AI announced a partnership with Agence France-Presse (AFP) to integrate the news agency's newswire into Le Chat. The integration will provide access to 2,300 daily stories in French, English, Spanish, Portuguese, German, and Arabic from AFP's network of 1,700 journalists.

Tencent released Hy-MT2, a family of multilingual translation models available in 1.8B, 7B, and 30B-A3B (MoE) sizes. All models support translation among 33 languages and follow translation instructions in multiple languages. The 1.8B model can be compressed to 440MB using 1.25-bit AngelSlim quantization.

Tencent has open-sourced Hy-MT2, a family of multilingual translation models available in 1.8B, 7B, and 30B-A3B parameter sizes. The models support translation across 33 languages and include extreme quantization down to 1.25-bit, reducing the 1.8B model to 440MB storage while increasing inference speed by 1.5x.

IBM released two new multilingual embedding models under Apache 2.0: a 97M-parameter compact model scoring 60.3 on MTEB Multilingual Retrieval (highest in its size class) and a 311M full-size model scoring 65.2. Both support 200+ languages with enhanced retrieval for 52 languages, handle 32K-token context (64x increase over predecessors), and include code retrieval across 9 programming languages.

Supertone has released Supertonic 3, a 99M-parameter text-to-speech model that runs entirely on-device using ONNX Runtime. The model expands language support from 5 to 31 languages compared to Supertonic 2, requires no GPU, and claims competitive accuracy against models 7-20x larger.

IBM released Granite Embedding 311M Multilingual R2, a 311-million parameter dense embedding model with 32,768-token context length and support for 200+ languages. The model scores 64.0 on Multilingual MTEB Retrieval (18 tasks), an 11.8-point improvement over its predecessor, and ships with ONNX and OpenVINO models for production deployment.

IBM released Granite 4.1 30B, a 30-billion parameter instruction-following model with a 131,072 token context window. The model scores 80.16 on MMLU 5-shot and 88.41 on HumanEval pass@1, with enhanced tool-calling capabilities following OpenAI's function definition schema.

IBM has released Granite 4.1 8B, an 8-billion-parameter decoder-only language model with a 131,072-token context window. The model supports 12 languages and costs $0.05 per million input tokens and $0.10 per million output tokens, available under the Apache 2.0 license.

IBM has released Granite 4.1-8B, an 8-billion parameter long-context model with a 131,072-token context window. The model achieves 85.37% on HumanEval and 73.84% on MMLU 5-shot, with enhanced tool-calling capabilities reaching 68.27% on BFCL v3. Released under Apache 2.0 license, it supports 12 languages.

TII UAE released QIMMA, an Arabic LLM leaderboard that validates benchmark quality before evaluating models. The validation pipeline, using Qwen3-235B and DeepSeek-V3 plus human review, discarded 3.1% of ArabicMMLU samples and found systematic quality issues across 14 benchmarks.

Liquid AI has released LFM2.5-VL-450M, a refreshed 450M-parameter vision-language model built on an updated LFM2.5-350M backbone. The model features a 32,768-token context window, supports 9 languages, handles native 512×512 pixel images, and adds bounding box prediction and function calling capabilities. Performance improvements span both vision and language benchmarks compared to its predecessor.

Microsoft's Bing team has open-sourced Harrier, a 27-billion-parameter embedding model that supports over 100 languages and features a 131,072-token context window. The model ranks first on the MTEB v2 multilingual benchmark, outperforming proprietary offerings from OpenAI and Amazon, and is available on Hugging Face under the MIT license.

Microsoft released the Harrier-OSS embedding model family, comprising three variants with 270M, 600M, and 27B parameters. The largest model achieves 74.3 on the Multilingual MTEB v2 benchmark. All models support 32,768 max tokens and multilingual inputs across 40+ languages.

Microsoft has released Harrier-OSS-v1, a family of multilingual text embedding models trained with contrastive learning and knowledge distillation. The 0.6B parameter variant achieves a 69.0 score on the Multilingual MTEB v2 benchmark with support for 32,768 token context windows and 45+ languages.

Google is expanding Search Live, its voice and camera-based AI search assistant, to more than 200 countries and territories with support for dozens of languages. The expansion is powered by Gemini 3.1 Flash Live, a new audio-focused model that Google claims offers faster response times and more natural conversations.

Google is rolling out Search Live globally, its conversational search feature powered by Gemini 3.1 Flash Live, which supports over 90 languages. Simultaneously, Google Translate's live headphones translation mode is launching on iOS after its Android debut, supporting over 70 languages across seven new countries.

Mistral AI released Voxtral-4B-TTS-2603, an open-weights text-to-speech model designed for production voice agents. The 4B-parameter model supports 9 languages, 20 preset voices, achieves 70ms latency at concurrency 1 on a single NVIDIA H200, and requires only 16GB GPU memory.

NVIDIA released Nemotron 3 Content Safety 4B, an open-source multimodal safety model designed to moderate content across text, images, and multiple languages. Built on Gemma-3 4B-IT with a 128K context window, the model achieved 84% average accuracy on multimodal safety benchmarks and supports over 140 languages through culturally-aware training data.

Descript has integrated OpenAI models to enable multilingual video dubbing at scale, optimizing translations for both semantic accuracy and speech timing to produce natural-sounding dubbed content. The system balances meaning preservation with practical constraints of dubbed audio synchronization.

NVIDIA has released Nemotron-3-Super-120B-A12B-NVFP4, a 120-billion parameter text generation model featuring a latent Mixture-of-Experts (MoE) architecture. The model supports 8 languages including English, French, Spanish, Italian, German, Japanese, and Chinese, and is available on Hugging Face with 8-bit quantization support through NVIDIA's ModelOpt toolkit.

NVIDIA has released Nemotron-3-Super-120B-A12B-BF16, a 120 billion parameter model designed for text generation and conversational tasks. The model employs a latent mixture-of-experts (MoE) architecture and supports multiple languages including English, French, Spanish, Italian, German, Japanese, and Chinese.

IBM has released Granite 4.0 1B Speech, a 1 billion parameter multilingual speech model designed for edge deployment. The model supports multiple languages and is optimized for devices with limited computational resources.

Liquid AI has released LFM2-24B-A2B, a 24-billion parameter mixture-of-experts model designed for text generation and conversational tasks. The model supports nine languages including English, Arabic, Chinese, French, German, Japanese, Korean, Spanish, and Portuguese.

Cohere Labs has released tiny-aya-global, a lightweight text generation model trained to support conversational tasks across 100+ languages. The model is available on Hugging Face under a CC-BY-NC-4.0 license and builds on the tiny-aya-base architecture.