Models

Voxtral Mini Transcribe 3 is Mistral's fast, cost-effective transcription model, with diarization, context biasing and speaker references in 40+ languages, built for high-volume audio processing.

by MistralOct 20, 202666K contextfrom $0.0035/M inputfrom $0.00/M output

Voxtral Mini Transcribe Realtime 3 is Mistral's streaming transcription model, with diarization, turn detection and configurable latency in 40+ languages, built for live, high-volume audio.

by MistralOct 20, 202666K contextfrom $0.0080/M inputfrom $0.00/M output

Voxtral TTS 3 is Mistral's text-to-speech model. It generates natural, expressive speech with controllable emotion and pace in 35+ languages and 50+ dialects, with preset voices and zero-shot voice cloning.

by MistralOct 20, 20264K contextfrom $13.00/M inputfrom $0.00/M output

Voxtral Small Transcribe 3 is Mistral's most accurate transcription model, with state-of-the-art multilingual transcription, diarization and context biasing in 40+ languages.

by MistralOct 20, 202666K contextfrom $0.0040/M inputfrom $0.00/M output

DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the `deepseek-flash` API model name. V4.1 Flash is a sparse mixture-of-experts model...

by DeepSeekOct 17, 20261M contextfrom $0.04/M inputfrom $0.19/M output2B tokens

Mistral Large 4 (Le Chonk), a general-purpose multimodal model available in public preview.

by MistralOct 6, 20261M context$0.70/M input$2.10/M output141 tokens

Mistral Large 4 is a frontier multimodal (text and image input) model from Mistral AI built for reasoning, coding, and agentic workloads.

by MistralOct 6, 20261M context$0.70/M input$2.10/M output

GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks.

by Z.aiAug 26, 20261M contextfrom $0.03/M inputfrom $0.09/M output43.4M tokens

This model always redirects to the latest GLM model from Z.ai.

by Z.aiAug 19, 20261M context$1.40/M input$4.40/M output

GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks.

by Z.aiAug 14, 20261M contextfrom $0.07/M inputfrom $1.40/M output3M tokens

Leanstral 1.5 is Mistral's model for formal mathematics and proof engineering in the Lean 4 proof assistant.

by MistralJun 30, 2026262K context$0.00/M input$0.00/M output33 tokens

Mistral Medium 3 5 is a Mistral AI model for coding, reasoning, instruction following, or multimodal language tasks.

by MistralApr 28, 2026262K context$1.50/M input$7.50/M output

Mistral Medium 3 5 is a Mistral AI model for coding, reasoning, instruction following, or multimodal language tasks.

by MistralApr 28, 2026262K contextfrom $0.75/M inputfrom $3.75/M output8 tokens

Voxtral Mini TTS is Mistral's text-to-speech model featuring zero-shot voice cloning and multilingual support. It converts text input into natural-sounding audio output.

by MistralApr 18, 20264K contextfrom $16.00/M inputfrom $0.00/M output

The mistral-small model is a cost-efficient reasoning model designed for low-latency workloads. It leverages the Mixtral-8X7B-v0.1 architecture, which incorporates a sparse mixture of experts with 12B active parameters. With improved reasoning capabilities, the model demonstrates a broad range of functionalities, including code generation and reasoning. It also supports multiple languages such as English, French, German, Italian, and Spanish, making it a versatile option for various linguistic tasks.

by MistralMar 16, 2026262K contextfrom $0.15/M inputfrom $0.60/M output

Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models, including strong reasoning, into a single system.

by MistralMar 16, 2026262K context$0.15/M input$0.60/M output8 tokens

Mistral's batch speech-to-text model (Voxtral Transcribe 2 family), with speaker diarization, context biasing and word-level timestamps across 13 languages.

by MistralFeb 4, 202616K contextfrom $0.0030/M inputfrom $0.00/M output

Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding. It is a 123B-parameter dense transformer model supporting a 256K context window. Devstral 2 supports exploring codebases and orchestrating changes across multiple files while maintaining architecture-level context. It tracks framework dependencies, detects failures, and retries with corrections—solving challenges like bug fixing and modernizing legacy systems. The model can be fine-tuned to prioritize specific languages or optimize for large enterprise codebases. It is available under a modified MIT license.

by MistralDec 9, 2025256K context$0.40/M input$2.00/M output1.2K tokens

Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding. It is a 123B-parameter dense transformer model supporting a 256K context window. Devstral 2 supports exploring codebases and orchestrating changes across multiple files while maintaining architecture-level context. It tracks framework dependencies, detects failures, and retries with corrections—solving challenges like bug fixing and modernizing legacy systems. The model can be fine-tuned to prioritize specific languages or optimize for large enterprise codebases. It is available under a modified MIT license.

by MistralDec 9, 2025262K context$0.40/M input$2.00/M output

Ministral 14b 2512 is a Mistral AI model for coding, reasoning, instruction following, or multimodal language tasks.

by MistralDec 2, 2025262K context$0.20/M input$0.20/M output