Ministral 3 3B review, pricing and limits

Ministral 3 3B fits mobile app developers, IoT engineers, and cost-sensitive pipeline teams who want on-device multimodal inference without a pricey API host. It handles high-volume classification, extraction, and translation across 11 supported languages, including Japanese, Korean, and Chinese. Teams needing verified math or logic benchmarks are better served by the Ministral 3 14B Reasoning variant instead.

Ministral 3 3B is Mistral AI's smallest open-weight multimodal model, pairing a compact language core with a 410M-parameter Vision Transformer encoder for image understanding. It supports a 256K token context window and runs on consumer GPUs or Apple Silicon under 8GB of VRAM, targeting on-device and edge inference.

Provider: Mistral AI · Family: Ministral 3

More about Mistral AI on HokAI

Context window: 256,000 tokens

Input modalities: text, image, tool-calls · Output: text, tool-calls

About Ministral 3 3B

Ministral 3 3B is the smallest member of Mistral AI's Ministral 3 family, released December 2, 2025. It uses a dense Transformer architecture combining 3.4 billion language decoder parameters with an integrated Vision Transformer (ViT) encoder for image understanding. Released under Apache 2.0, it is freely usable for commercial self-hosting. In Mistral's product lineup it sits below the Ministral 3 8B and 14B models and far below the Mistral Large 3 MoE at 675B total parameters. The design priority is edge and on-device deployment where 8GB of VRAM or less is available.

Mistral has not published a standalone MMLU-Pro or GPQA score for the 3B variant. At the 8B scale in the same family, the base model scores 70.6% on MMLU-Pro; at 14B the base scores 74.2%. Llama 3.2 3B scores approximately 58% on standard MMLU, placing the Ministral 3 3B well above it on general knowledge tasks, though Mistral has not published a direct verified comparison. The Ministral 3 8B Reasoning variant achieves 78.7% on AIME 2025 and 66.8% on GPQA Diamond, giving a sense of the family's ceiling; the 3B reasoning results have not been separately released.

The context window is large for a model this size, letting it hold long documents or extended conversations without external retrieval. It uses Grouped Query Attention (GQA) with 32 query heads and 8 key-value heads, which reduces KV cache memory at long contexts. Mistral has not published a formal needle-in-haystack long-context recall evaluation for the 3B variant, and maximum output tokens per API call are not separately documented; most use cases stay under 4,096 output tokens.

Ministral 3 3B accepts text and image inputs through its integrated ViT encoder. It can describe images, answer visual questions about photographs and charts, and extract printed text from images. Output is text only; image generation is not supported. Function calling and structured JSON output are available via Mistral's OpenAI-compatible tool-use schema. Audio and video inputs are not supported. The model is multilingual across European languages (French, German, Spanish, Italian, Portuguese, Dutch) and East Asian languages (Japanese, Korean, Chinese) alongside English.

API access runs through Mistral's la Plateforme at api.mistral.ai with model ID ministral-3-3b-latest or ministral-3-3b-2512. Weights are available on Hugging Face at mistralai/Ministral-3-3B-Instruct-2512 (instruct), mistralai/Ministral-3-3B-Base-2512 (base), and mistralai/Ministral-3-3B-Reasoning-2512 (reasoning). In FP8 format the model fits in 8GB of VRAM; Q4 quantization via GGUF reduces memory below 4GB for deployment on Apple M-series chips and RTX 30/40 series consumer GPUs. The recommended self-hosting framework is vLLM; llama.cpp and LM Studio support the model via standard GGUF conversion.

Safety alignment uses supervised fine-tuning plus RLHF on the instruct variant. Mistral's approach is lighter on refusals than comparable Anthropic or OpenAI models, consistent with the company's position that safety guardrails should be configurable by the deploying organization. No separate system card or red team report has been published for the 3B model. API inputs via la Plateforme are subject to Mistral's abuse monitoring retention policy; inputs are not used for model training. Fully air-gapped self-hosted deployments have no data exposure.

Teams that need strong math or code generation should step up to the Ministral 3 14B Reasoning variant, which scores 85% on AIME 2025. Real-time voice assistant builders should avoid this model because there is no native audio input. For very long agentic coding loops, Mistral Large 3 or the Ministral 3 14B are better suited.

The Ministral 3 3B was trained on a multilingual dataset with a knowledge cutoff in or around late 2025, consistent with other Ministral 3 family models. Mistral AI is headquartered in Paris and operates as a GDPR-compliant European company; API traffic via la Plateforme routes through European infrastructure by default. Enterprise data handling agreements and zero-retention options are available through Mistral's commercial channel.

Pricing

$0.10 per 1M input tokens and $0.10 per 1M output tokens via la Plateforme. No cached-input rate published. Self-hosted under the Apache license incurs only compute cost.

What a real job costs

Budgets: 20-page PDF = 30k in / 1k out · Support reply = 2k in / 300 out · Coding agent run = 200k in / 20k out. Computed from the vendor's per-token prices at render time; cached-input discounts are not applied.

Key Features

• 256K Context Window: Supports a 256K token context window, double the 128K standard for Llama 3.2 3B at the same parameter scale, enabling full-book or long-document processing at edge scale.
• Integrated Vision Encoder: A 410M-parameter ViT encoder processes images natively, enabling visual QA, chart reading, and OCR without a separate vision model.
• Open-Source Apache License: Released under Apache 2.0, the license permits commercial use, modification, and redistribution with no royalties or usage restrictions.
• 8GB FP8 Footprint: Runs in 8GB of VRAM in FP8, or under 4GB with Q4 quantization, supporting consumer RTX GPUs and Apple M-series chips.
• Multilingual by Design: First-class support for 11 languages including French, German, Spanish, Japanese, Korean, and Chinese, not just English fine-tuning.

Pros

• Fits in 4GB Q4 VRAM, the smallest footprint of any multimodal instruction-following model in its quality tier.
• Cheaper flat per-token pricing than GPT-4o Mini for both input and output, ideal for high-volume pipelines.
• The Apache-licensed, open-source release allows fully air-gapped self-hosted deployment with no API dependency or per-token fees.

Cons

• No published MMLU-Pro or GPQA benchmark for the 3B specifically; users must benchmark on their own tasks.
• No native audio input; voice applications require a separate speech-to-text step.
• Lighter safety alignment than larger Mistral models; deployers must build their own content guardrails.

Benchmarks

• AA blended price: $0.1/M cited: Artificial Analysis — Price per 1M tokens at a 3:1 input to output blend, as listed by Artificial Analysis.

A benchmark is an exam, not the job. Scores transfer unevenly between tasks, so weigh the one closest to your workload and read every figure with its source.

Visit Ministral 3 3B Official Page