Ministral 3 3B Reasoning 2512 — Hardware Requirements & GPU Compatibility
ReasoningFunctions
Ministral 3 3B Reasoning 2512 is a 3.8B-parameter open language model from Mistral AI in the Mistral family. It supports a context window of up to 262,144 tokens. At Q4_K_M it needs about 2.77 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 64K)Best on 24 GB (Q4_K_M · ~237 tok/s · 256K)Macs from 8 GB (Q4_K_M · ~26 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Ministral 3 3B Reasoning 2512
One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.
Recommended
ollama run hf.co/MaziyarPanahi/Ministral-3-3B-Reasoning-2512-GGUF:Q4_K_MQ4_K_M fits fully on the RTX 4090 · 20.1 GB free · 16K context
Download: Ministral-3-3B-Reasoning-2512.Q4_K_M.gguf2.15 GBMaziyarPanahi
Ways to runOther ways
each opens step-by-stepRun in cloud
Your own private copy · Q4_K_M · 24 GB GPU
- The easiest way to try this model
- Running in a few minutes, easy setup
- Chat right here or connect any app
$5+ bonus for new users
Runs on RTX 3060 12GB (8 GB headroom) · Q4_K_M
- Generation speed
- ~85 tok/s
- generation speed
- Cost per 1M output tokens
- $0.20
- per 1M output tokens
How Much VRAM Does Ministral 3 3B Reasoning 2512 Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 2.1 GB | 4.6 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 256K context · ~312 tok/s | |
| 2.3 GB | 4.8 GB | 3-bit medium — 3-bit medium quantization | Fits: 256K context · ~280 tok/s | |
| 2.4 GB | 4.9 GB | 3-bit large — 3-bit large quantization | Fits: 256K context · ~269 tok/s | |
| 2.8 GB | 5.2 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 256K context · ~237 tok/s | |
| 3.2 GB | 5.7 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 256K context · ~204 tok/s | |
| 3.6 GB | 6.1 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 128K context · ~180 tok/s | |
| 4.3 GB | 6.8 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 128K context · ~152 tok/s | |
| 8.2 GB | 10.6 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 128K context · ~80 tok/s |
Selected Q4_K_M
4-bit medium quantization3.9GBat 16K context
- Weights
- 2.3 GB
- KV cache
- 1.3 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 20.1 GB free · ~237 tok/s
≥50 tok/s needs ≥213 GB/s · RTX 5090 ~421, RTX 3090 Ti ~237
Which Hardware Can Run Ministral 3 3B Reasoning 2512?
Q4_K_M · 2.8 GBMinistral 3 3B Reasoning 2512 (Q4_K_M) requires 2.8 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 4+ GB is recommended. Using the full 262K context window can add up to 20.8 GB, bringing total usage to 23.6 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run Ministral 3 3B Reasoning 2512, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 4.1 GB free · up to 64K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~178Simulate output on RTX 3080RTX 2080 Ti~145Simulate output on RTX 2080 TiRTX 3070 Ti~143Simulate output on RTX 3070 TiM3*8 GB~26Simulate output on iMac 24" M3 (8 GB)Orin Nano~24Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~17Simulate output on iPhone 1720 GB
fits · 16.1 GB free · up to 128K ctx
1 GPU
RX 7900 XT~173Simulate output on RX 7900 XT32–40 GB
fits · 28.1 GB free · up to 256K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~421Simulate output on RTX 5090M5 Max36 GB~116Simulate output on Mac Studio M5 Max (36 GB)M532 GB~39Simulate output on MacBook Air 13" M5 (32 GB)Orin~48Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~32Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Alternatives to Ministral 3 3B Reasoning 2512
Mistral family · 35 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Spark X2.6 4BXHToken | — | 4.3 GB | ~221 | same size |
| NVIDIA Nemotron Labs Dory1.0 4B BF16NVIDIA | — | 5.7 GB | ~214 | same size |
| Granite Vision 4.1 4BIBM | — | 4 GB | ~228 | same size |
| Qwen3 4B Instruct 2507Alibaba | — | 4.2 GB | ~226 | same size |
| MiniCPM V 4OpenBMB | — | 3.1 GB | ~236 | same size |
| Phi 4 Mini ReasoningMicrosoft | — | 4.8 GB | ~228 | same size |
Specifications
- Publisher
- Mistral AI
- Family
- Mistral
- Parameters
- 3.8B
- Architecture
- Mistral3ForConditionalGeneration
- Context Length
- 262,144 tokens
- Vocabulary Size
- 131,072
- Release Date
- 2025-10-31
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does Ministral 3 3B Reasoning 2512 need?
Ministral 3 3B Reasoning 2512 requires 2.8 GB of VRAM at Q4_K_M, or 8.2 GB at BF16. Full 262K context adds up to 20.8 GB (23.6 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 3.8B × 4.8 bits ÷ 8 = 2.3 GB
KV Cache + Overhead ≈ 0.5 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 21.3 GB (at full 262K context)
VRAM usage by quantization
What's the best quantization for Ministral 3 3B Reasoning 2512?
For Ministral 3 3B Reasoning 2512, Q4_K_M (2.8 GB) offers the best balance of quality and VRAM usage. Q5_K_M (3.2 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 2.1 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Ministral 3 3B Reasoning 2512 on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run Ministral 3 3B Reasoning 2512. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 2.8 GB of usable unified memory (RAM minus macOS overhead).
Can I run Ministral 3 3B Reasoning 2512 locally?
Yes — Ministral 3 3B Reasoning 2512 can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Ministral 3 3B Reasoning 2512?
At Q4_K_M, Ministral 3 3B Reasoning 2512 can reach ~1733 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~237 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 2.77 × 0.65 = ~1877 tok/s
Estimated speed at Q4_K_M (2.8 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Ministral 3 3B Reasoning 2512?
At Q4_K_M, the download is about 2.31 GB. The full-precision BF16 version is 7.70 GB. The smallest option (Q2_K) is 1.64 GB.
Which GPUs can run Ministral 3 3B Reasoning 2512?
52 consumer GPUs can run Ministral 3 3B Reasoning 2512 at Q4_K_M. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 52 GPUs have plenty of headroom for comfortable inference.
Which devices can run Ministral 3 3B Reasoning 2512?
163 devices with unified memory can run Ministral 3 3B Reasoning 2512 at Q4_K_M (2.8 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.