new·The score now tells you which way it movedA brain's exam only ever grows: its own material writes questions, and so does every question a real caller asked and did not get answered. The score is a percentage over that growing set, so a brain that learned more could post a smaller number — and this week three did. One of them answered two MORE questions than the week before and showed eighteen points less. Printed as a single percentage, that reads as decline to a reader and as punishment to anyone who contributes material.all news →
mozg.beta
Sign in

AI SDK · Providers · all subjects

mistral/models

9 notes, read out of this brain and free to use. Each one was extracted from a source and is re-checked against its exam.

Mistral language model instantiation

Mistral language models are created using the provider instance with a model ID as the first argument, e.g., `mistral('mistral-large-latest')`. Models call the Mistral chat API.

Mistral reasoning models

Mistral offers reasoning models including magistral-small-2507 and magistral-medium-2507 that provide step-by-step thinking capabilities. These models return structured reasoning content with reasoningText available via the reasoningText property in the result.

Mistral transcription model instantiation

Mistral transcription models are created using the `.transcription()` factory method with a model ID, e.g., `mistral.transcription('voxtral-mini-latest')`. These models call the Mistral audio transcription API.

Mistral transcription model capabilities table

Transcription model capabilities: voxtral-mini-latest (Timestamps: yes, Diarization: yes, Context Bias: yes).

Mistral speech model instantiation

Mistral speech models are created using the `.speech()` factory method with a model ID, e.g., `mistral.speech('voxtral-mini-tts-2603')`. These models call the Mistral speech API.

Mistral speech model capabilities table

Speech model capabilities: voxtral-mini-tts-2603 (Saved Voices: yes, Reference Audio: yes, Non-Streaming: yes).

Mistral embedding model instantiation

Mistral embedding models are created using the `.embedding()` factory method with a model ID, e.g., `mistral.embedding('mistral-embed')`. These models call the Mistral embeddings API.

Mistral embedding model capabilities table

Embedding model capabilities: mistral-embed (Default Dimensions: 1024).

Mistral language model capabilities

Mistral language model capabilities table: - pixtral-large-latest: Image Input (yes), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - mistral-large-latest: Image Input (no), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - mistral-medium-latest: Image Input (no), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - mistral-medium-3: Image Input (no), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - mistral-medium-2508: Image Input (no), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - mistral-medium-2505: Image Input (no), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - mistral-medium-3.5: Image Input (no), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - mistral-small-latest: Image Input (no), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - magistral-small-2507: Image Input (no), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - magistral-medium-2507: Image Input (no), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - magistral-small-2506: Image Input (no), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - magistral-medium-2506: Image Input (no), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - ministral-3b-latest: Image Input (no), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - ministral-8b-latest: Image Input (no), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - pixtral-12b-2409: Image Input (yes), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - open-mistral-7b: Image Input (no), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - open-mixtral-8x7b: Image Input (no), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes) - open-mixtral-8x22b: Image Input (no), Object Generation (yes), Tool Usage (yes), Tool Streaming (yes)

Give your agent this brain