North Micro Vision Instruct — Hardware Requirements & GPU Compatibility
VisionChat
North Micro Vision Instruct is Cohere's 2.5-billion-parameter vision-language model, built to handle native-resolution images alongside text in multi-turn conversations. It pairs a custom vision encoder with a small in-house language model, tuned for document understanding, chart and table reading, spatial reasoning, and visual grounding across a dozen-plus languages. Its small size suits local deployment on modest consumer GPUs once quantized. The model supports a very large 500,000 token context window, well beyond most models its size, useful for many-page documents or long image-heavy conversations. It is released under the Apache 2.0 license, allowing unrestricted commercial and research use. Published in August 2026 as part of Cohere's North family, it accepts inputs at resolutions up to roughly an A4 page at 200 dpi, aimed at document-heavy enterprise use.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 32K)Best on 24 GB (Q4_K_M · ~323 tok/s · 128K)Macs from 8 GB (Q4_K_M · ~35 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run North Micro Vision Instruct
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
vllm serve CohereLabs/North-Micro-Vision-Instruct --max-model-len 16384vLLM repo with its own quantisation
Download: Weights on Hugging Face
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3060 12GB (8 GB headroom) · Q4_K_M
- Generation speed
- ~115 tok/s
- generation speed
- Cost per 1M output tokens
- $0.14
- per 1M output tokens
How Much VRAM Does North Micro Vision Instruct Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 1.6 GB | 5.1 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 128K context · ~412 tok/s | |
| 1.8 GB | 5.3 GB | 3-bit medium — 3-bit medium quantization | Fits: 128K context · ~374 tok/s | |
| 2.0 GB | 5.6 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 128K context · ~323 tok/s | |
| 2.3 GB | 5.8 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 128K context · ~284 tok/s | |
| 2.6 GB | 6.1 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 128K context · ~254 tok/s | |
| 3.0 GB | 6.5 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 128K context · ~217 tok/s | |
| 5.5 GB | 9 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 128K context · ~119 tok/s |
Selected Q4_K_M
4-bit medium quantization3.7GBat 16K context
- Weights
- 1.5 GB
- KV cache
- 1.9 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 20.3 GB free · ~323 tok/s
≥50 tok/s needs ≥156 GB/s · RTX 5090 ~574, RTX 3090 Ti ~323
Which Hardware Can Run North Micro Vision Instruct?
Q4_K_M · 2.0 GBNorth Micro Vision Instruct (Q4_K_M) requires 2.0 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 3+ GB is recommended. Using the full 500K context window can add up to 57.1 GB, bringing total usage to 59.1 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run North Micro Vision Instruct, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 4.3 GB free · up to 32K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~243Simulate output on RTX 3080RTX 2080 Ti~197Simulate output on RTX 2080 TiRTX 3070 Ti~195Simulate output on RTX 3070 TiM3*8 GB~35Simulate output on iMac 24" M3 (8 GB)Orin Nano~33Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~24Simulate output on iPhone 1720 GB
fits · 16.3 GB free · up to 128K ctx
1 GPU
RX 7900 XT~237Simulate output on RX 7900 XT32–40 GB
fits · 28.3 GB free · up to 256K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~574Simulate output on RTX 5090M5 Max36 GB~159Simulate output on Mac Studio M5 Max (36 GB)M532 GB~53Simulate output on MacBook Air 13" M5 (32 GB)Orin~66Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~43Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| WeVisOne VL 2BTencent | — | 3.6 GB | ~328 | same size |
| MiniCPM5 2BOpenBMB | — | 2.5 GB | ~345 | same size |
| Granite 3.1 2B InstructIBM | — | 3.2 GB | ~329 | same size |
| Gemma 2 2B ITGoogle | — | 2.5 GB | ~379 | same size |
| Ouro 2.6B ThinkingByteDance | — | 12.1 GB | ~102 | same size |
| EXAONE 3.5 2.4B InstructLGAI-EXAONE | — | 3.4 GB | ~412 | same size |
Specifications
- Publisher
- Cohere
- Family
- North
- Parameters
- 2.5B
- Architecture
- CohereCompassForConditionalGeneration
- Context Length
- 500,000 tokens
- Vocabulary Size
- 262,144
- Release Date
- 2026-08-10
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does North Micro Vision Instruct need?
North Micro Vision Instruct requires 2.0 GB of VRAM at Q4_K_M, or 5.5 GB at BF16. Full 500K context adds up to 57.1 GB (59.1 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 2.5B × 4.8 bits ÷ 8 = 1.5 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 ≈ 57.6 GB (at full 500K context)
VRAM usage by quantization
What's the best quantization for North Micro Vision Instruct?
For North Micro Vision Instruct, Q4_K_M (2.0 GB) offers the best balance of quality and VRAM usage. Q5_K_M (2.3 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 1.6 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run North Micro Vision Instruct on a Mac?
Yes — Mac Mini M2 (8 GB) and 145 other Macs can run North Micro Vision Instruct. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 2.0 GB of usable unified memory (RAM minus macOS overhead).
Can I run North Micro Vision Instruct locally?
Yes — North Micro Vision Instruct can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is North Micro Vision Instruct?
At Q4_K_M, North Micro Vision Instruct can reach ~2365 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~323 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 ÷ 2.03 × 0.65 = ~2562 tok/s
Estimated speed at Q4_K_M (2.0 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of North Micro Vision Instruct?
At Q4_K_M, the download is about 1.49 GB. The full-precision BF16 version is 4.97 GB. The smallest option (Q2_K) is 1.06 GB.
Which GPUs can run North Micro Vision Instruct?
52 consumer GPUs can run North Micro Vision Instruct 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 North Micro Vision Instruct?
163 devices with unified memory can run North Micro Vision Instruct at Q4_K_M (2.0 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.