Discover AI models for every task
Showing 1-17 of 17 models
by BAAI
BAAI BGE-M3 is a versatile multilingual embedding model supporting dense, sparse, and multi-vector retrieval in a unified architecture. Handles 100+ languages with strong cross-lingual capabilities and flexible retrieval modes for different use cases. Features hybrid retrieval combining dense embeddings for semantic similarity, sparse representations for lexical matching, and multi-vector approaches for fine-grained relevance. Ideal for multilingual search engines, hybrid retrieval systems, and complex information retrieval scenarios requiring multiple matching strategies.
BAAI BGE Multilingual Gemma2 is a multilingual dense retrieval embedding model built on Gemma 2 architecture, supporting 100+ languages for cross-lingual semantic search and retrieval. Delivers strong performance across diverse language families including English, Chinese, Spanish, Arabic, Hindi, and many more. Ideal for multilingual search systems, cross-lingual document retrieval, international content recommendation, and global knowledge bases. Trained on large-scale multilingual data with balanced language representation.
BAAI BGE Large EN V1.5 is a state-of-the-art English dense retrieval embedding model with 1024-dimensional embeddings and 512 token sequence length. Achieves 64.23 average on MTEB leaderboard across 56 tasks with 54.29 on retrieval. Pre-trained with RetroMAE and fine-tuned on large-scale contrastive learning data. V1.5 improvements include better similarity distribution and flexible usage without query instructions. Ideal for semantic search, document retrieval, re-ranking pipelines, and sentence similarity tasks. Production-ready with 3.4M+ downloads/month.
BAAI BGE EN ICL is an in-context learning enabled English embedding model supporting dynamic query understanding through examples. Features innovative ICL approach allowing users to provide examples to guide retrieval behavior without retraining. Excels at domain-specific retrieval tasks where query intent can be demonstrated through few-shot examples. Ideal for specialized search applications, adaptive retrieval systems, and scenarios requiring customizable semantic understanding. Released July 2024 with state-of-the-art ICL embedding capabilities.
by MiniMax
MiniMax M2 is a powerful MoE model with 200B total / 10B active parameters. Optimized for reasoning and coding tasks with excellent performance in multilingual scenarios. Features 128K context window and efficient inference.
MiniMax M2.5 is a state-of-the-art reasoning MoE model with 229B total / 10B active parameters. Extensively trained with reinforcement learning across 200,000+ real-world environments, achieving SOTA performance in coding (80.2% SWE-Bench Verified), agentic tool use, search, and office productivity tasks. Features 197K context window, efficient MoE inference, and strong multilingual support.
by OpenAI
OpenAI Whisper Large V3 is a state-of-the-art automatic speech recognition model with 1550M parameters supporting 99 languages. Achieves 10-20% WER reduction compared to V2, trained on 1M hours weakly labeled + 4M hours pseudo-labeled audio. Features 128 Mel frequency bins (increased from 80), improved robustness to accents and background noise, and new Cantonese language support. Supports speech transcription and speech-to-English translation with sentence and word-level timestamps. Optimized with torch.compile for 4.5x speedup. Ideal for accessibility tools, multilingual transcription, and enterprise ASR applications.
MiniMax M2.1 is a state-of-the-art MoE model with 230B total / 10B active parameters, optimized for agentic coding and complex multi-step workflows. Excels at multilingual programming, tool use, and long-horizon planning. Matches Claude Sonnet 4.5 on code benchmarks and exceeds it in multilingual scenarios. Features 196K context window with FP8 efficiency. Released under Modified-MIT license for commercial use.
by Mistral
Mistral Small 4 is a 119B-parameter Mixture-of-Experts model (128 experts, 4 active per token, 6.5B active parameters) that unifies instruct, reasoning, and coding capabilities into a single multimodal model. It accepts text and image inputs, supports function calling, structured outputs, and configurable reasoning effort (none for fast responses, high for deep step-by-step reasoning). With a 256K context window and Apache 2.0 license, it delivers 40% lower latency and 3x higher throughput compared to Mistral Small 3.
MiniMax M2.7 is a MiniMaxM2 architecture model with an undisclosed parameter count, optimized for advanced reasoning and agentic workflows. It excels at complex tool use, self‑evolution, and professional software engineering tasks, achieving a 66.6% medal rate on MLE Bench Lite and a 56.2% score on SWE Bench Pro. The model also attains an ELO of 1495 on GDPval‑AA, surpassing other open‑weight models. Available under an Other license.
Mistral Small 3.2 24B Instruct is a multimodal instruction-tuned model supporting both vision and text with 24B parameters and 128K context. Major improvements over 3.1 include better instruction following (84.78%), 2x reduction in repetition errors, and robust function calling. Achieves 65.33% on Wildbench v2, 43.1% on Arena Hard v2, 92.90% on HumanEval Pass@5. Vision benchmarks: 87.4% ChartQA, 94.86% DocVQA, 62.50% MMMU. Supports up to 10 images per prompt with integrated vision-based function calling.
by intfloat
intfloat E5-Mistral-7B-Instruct is a state-of-the-art instruction-following embedding model built on Mistral 7B architecture. Combines strong language understanding from Mistral with specialized embedding training for retrieval tasks. Features instruction-based embedding generation allowing natural language queries to guide semantic search. Excels at complex retrieval scenarios, multi-hop reasoning in document search, and instruction-guided similarity tasks. Provides significantly improved zero-shot retrieval performance compared to traditional embedding models.
OpenAI Whisper Large V3 Turbo is an optimized variant of Whisper V3 with significantly faster inference while maintaining high accuracy across 99 languages. Features architectural optimizations for reduced latency including faster encoder-decoder inference and efficient attention mechanisms. Delivers near-V3 accuracy with 2-3x speed improvement, ideal for real-time transcription applications, live subtitling, and high-throughput ASR workloads. Supports full multilingual capabilities, timestamps, and speech translation to English. Perfect for production deployments requiring both quality and speed.
by Sentence Transformers
Sentence Transformers Paraphrase Multilingual MPNet is a multilingual sentence embedding model based on MPNet architecture, supporting 50+ languages for cross-lingual semantic similarity and paraphrase detection. Trained on large-scale paraphrase datasets across multiple languages enabling strong cross-lingual transfer. Ideal for multilingual paraphrase detection, semantic textual similarity, cross-lingual search, and international content deduplication. Provides balanced performance across diverse language families with proven track record in sentence-transformers ecosystem.
by Qwen
Image generation model from the Qwen series with advanced text rendering and precise image editing capabilities.
intfloat Multilingual E5 Large is a powerful multilingual dense retrieval embedding model supporting 100+ languages with strong cross-lingual capabilities. Features 1024-dimensional embeddings optimized for semantic search, document retrieval, and text similarity across diverse language families. Pre-trained on large-scale multilingual data with contrastive learning for robust cross-lingual transfer. Ideal for international search systems, multilingual document retrieval, and global content recommendation platforms requiring high-quality semantic understanding.
intfloat Multilingual E5 Large Instruct is an instruction-tuned multilingual embedding model combining strong cross-lingual capabilities with instruction-following for guided retrieval. Supports 100+ languages with natural language instructions to customize embedding behavior. Features enhanced zero-shot retrieval performance through instruction-based query understanding. Ideal for complex multilingual search scenarios, domain-specific retrieval tasks, and applications requiring adaptive semantic understanding across languages.