Cogito v1 — local AI model by Deep Cogito
Written by Jakub Rusinowski · Last updated
Deep Cogito's hybrid reasoning model family released March 2026. Unique 'think-or-not-think' architecture lets the model decide at inference time whether to use chain-of-thought reasoning based on query difficulty — saving compute on simple tasks while using deep reasoning for hard ones. Apache 2.0 licensed with sizes from 3B to 70B, all runnable on consumer hardware.
Variants
The smallest Cogito v1 variant needs about 3 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache.
| Model | VRAM at Q4 | VRAM | Context | Run it |
|---|---|---|---|---|
| Cogito v1 3B → 3B | ~2.6 GB | 32,000 | ollama run cogito:3b | |
| Cogito v1 8B → 8B | ~5.6 GB | 32,000 | ollama run cogito:8b | |
| Cogito v1 14B → 14B | ~9.3 GB | 64,000 | ollama run cogito:14b | |
| Cogito v1 32B → 32B | ~20.1 GB | 64,000 | ollama run cogito:32b | |
| Cogito v1 70B → 70B | ~43.1 GB | 128,000 | ollama run cogito:70b |
Memory is quantized weights plus overhead at Q4_K_M, from the same engine as the GPU & VRAM checker.
How to run Cogito v1 locally
Install Ollama, then pull the tag.
ollama run cogito:3bPick a size above for its own VRAM figure, speed estimate and install command.
Licence
Commercial use permitted. No usage restrictions beyond attribution.
Applies to: Cogito v1 3B, Cogito v1 8B, Cogito v1 14B, Cogito v1 32B, Cogito v1 70BRecommended GPU
The cheapest catalogued GPU that runs Cogito v1 locally (min 3 GB VRAM) is the Intel Arc B570 (10 GB).
Can I run Cogito v1 on my GPU?
- Cogito v1 on AMD Radeon RX 7800 XT
- Cogito v1 on AMD Radeon RX 7900 XT
- Cogito v1 on AMD Radeon RX 7900 XTX
- Cogito v1 on AMD Radeon RX 9060 XT 16GB
- Cogito v1 on AMD Radeon RX 9060 XT 8GB
- Cogito v1 on AMD Radeon RX 9070
- Cogito v1 on AMD Radeon RX 9070 XT
- Cogito v1 on AMD Ryzen AI Max+ 395
- Cogito v1 on Apple M1
- Cogito v1 on Apple M1 Max
- Cogito v1 on Apple M1 Pro
- Cogito v1 on Apple M2 Max
- Cogito v1 on Apple M2 Pro
- Cogito v1 on Apple M3 Max
- Cogito v1 on Apple M3 Pro
- Cogito v1 on Apple M4
- Cogito v1 on Apple M5
- Cogito v1 on Apple M5 Pro
- Cogito v1 on Intel Arc B570
- Cogito v1 on Intel Arc B580
- Cogito v1 on NVIDIA A100 80GB (PCIe)
- Cogito v1 on NVIDIA GeForce RTX 3060 (12GB)
- Cogito v1 on NVIDIA GeForce RTX 3070
- Cogito v1 on NVIDIA GeForce RTX 3070 Ti
- Cogito v1 on NVIDIA GeForce RTX 3080 (10GB)
- Cogito v1 on NVIDIA GeForce RTX 3090
- Cogito v1 on NVIDIA GeForce RTX 4060
- Cogito v1 on NVIDIA GeForce RTX 4060 Ti 16GB
- Cogito v1 on NVIDIA GeForce RTX 4070
- Cogito v1 on NVIDIA GeForce RTX 4070 Super
- Cogito v1 on NVIDIA GeForce RTX 4070 Ti
- Cogito v1 on NVIDIA GeForce RTX 4070 Ti Super
- Cogito v1 on NVIDIA GeForce RTX 4080
- Cogito v1 on NVIDIA GeForce RTX 4080 Super
- Cogito v1 on NVIDIA GeForce RTX 4090
- Cogito v1 on NVIDIA GeForce RTX 5060
- Cogito v1 on NVIDIA GeForce RTX 5060 Ti 16GB
- Cogito v1 on NVIDIA GeForce RTX 5060 Ti 8GB
- Cogito v1 on NVIDIA GeForce RTX 5070
- Cogito v1 on NVIDIA GeForce RTX 5070 Ti
- Cogito v1 on NVIDIA GeForce RTX 5080
- Cogito v1 on NVIDIA GeForce RTX 5090
- Cogito v1 on NVIDIA H100 80GB (PCIe)
- Cogito v1 on NVIDIA L40S
- Cogito v1 on NVIDIA RTX 6000 Ada Generation
- Cogito v1 on NVIDIA RTX PRO 6000 Blackwell
Cogito v1 — frequently asked questions
How much VRAM does Cogito v1 need?
Cogito v1 needs about 3 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: Cogito v1 3B (3 GB, Q4_K_M); Cogito v1 8B (6 GB, Q4_K_M); Cogito v1 14B (9 GB, Q4_K_M); Cogito v1 32B (20 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Can I run Cogito v1 on an RTX 4090 (24 GB)?
Yes — Cogito v1 runs on an RTX 4090 (24 GB) and other 24 GB cards such as the RTX 3090. Smaller variants also fit comfortably on 8–16 GB GPUs at Q4_K_M.
What quantization should I use for Cogito v1?
Q4_K_M is the best balance of quality and VRAM for Cogito v1 in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
How do I run Cogito v1 with Ollama?
Install Ollama, then run: ollama run cogito:3b. This downloads Cogito v1 and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.