Jan v3 4B Base Instruct GGUF — Hardware Requirements & GPU Compatibility
ChatCode
Jan v3 4B Base Instruct is a 4-billion-parameter model from the Jan AI project, provided in GGUF format for local deployment. Built on the Menlo Jan-v3-4B architecture, it is designed as a capable small assistant for both code and general chat, balancing helpfulness with a compact size that runs on modest consumer hardware. This model is a solid option for users exploring the Jan ecosystem or anyone who wants a lightweight local assistant that handles coding questions and everyday conversation in a single package. Its small parameter count keeps memory usage low, making it viable on laptops and entry-level desktop GPUs alike.
315.6K downloads
48 likesSpecifications
- Publisher
- janhq
- Parameters
- 4B
- Release Date
- 2026-01-28
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does Jan v3 4B Base Instruct GGUF Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q3_K_S | 3.50 | 1.9 GB | — | 1.75 GB | 3-bit small quantization |
| Q3_K_M | 3.90 | 2.1 GB | — | 1.95 GB | 3-bit medium quantization |
| Q4_0 | 4.00 | 2.2 GB | — | 2.00 GB | 4-bit legacy quantization |
| Q3_K_L | 4.10 | 2.3 GB | — | 2.05 GB | 3-bit large quantization |
| Q4_K_S | 4.50 | 2.5 GB | — | 2.25 GB | 4-bit small quantization |
| Q4_1 | 4.50 | 2.5 GB | — | 2.25 GB | 4-bit legacy quantization with offset |
| Q4_K_M | 4.80 | 2.6 GB | — | 2.40 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_0 | 5.00 | 2.8 GB | — | 2.50 GB | 5-bit legacy quantization |
| Q5_1 | 5.50 | 3.0 GB | — | 2.75 GB | 5-bit legacy quantization with offset |
| Q5_K_S | 5.50 | 3.0 GB | — | 2.75 GB | 5-bit small quantization |
| Q5_K_M | 5.70 | 3.1 GB | — | 2.85 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 3.6 GB | — | 3.30 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 4.4 GB | — | 4.00 GB | 8-bit quantization, near-lossless |
Which GPUs Can Run Jan v3 4B Base Instruct GGUF?
Q4_K_M · 2.6 GBJan v3 4B Base Instruct GGUF (Q4_K_M) requires 2.6 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 4+ GB is recommended. 35 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.
Runs great
— Plenty of headroomWhich Devices Can Run Jan v3 4B Base Instruct GGUF?
Q4_K_M · 2.6 GB33 devices with unified memory can run Jan v3 4B Base Instruct GGUF, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
Runs great
— Plenty of headroomRelated Models
Frequently Asked Questions
- How much VRAM does Jan v3 4B Base Instruct GGUF need?
Jan v3 4B Base Instruct GGUF requires 2.6 GB of VRAM at Q4_K_M, or 4.4 GB at Q8_0.
VRAM = Weights + KV Cache + Overhead
Weights = 4B × 4.8 bits ÷ 8 = 2.4 GB
KV Cache + Overhead ≈ 0.2 GB (at 2K context + ~0.3 GB framework)
VRAM usage by quantization
Q4_K_M2.6 GB- What's the best quantization for Jan v3 4B Base Instruct GGUF?
For Jan v3 4B Base Instruct GGUF, Q4_K_M (2.6 GB) offers the best balance of quality and VRAM usage. Q5_0 (2.8 GB) provides better quality if you have the VRAM. The smallest option is Q3_K_S at 1.9 GB.
VRAM requirement by quantization
Q3_K_S1.9 GB~77%Q3_K_L2.3 GB~86%Q4_K_M ★2.6 GB~89%Q5_02.8 GB~90%Q5_K_S3.0 GB~92%Q8_04.4 GB~99%★ Recommended — best balance of quality and VRAM usage.
- Can I run Jan v3 4B Base Instruct GGUF on a Mac?
Jan v3 4B Base Instruct GGUF requires at least 1.9 GB at Q3_K_S, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run Jan v3 4B Base Instruct GGUF locally?
Yes — Jan v3 4B Base Instruct GGUF can run locally on consumer hardware. At Q4_K_M quantization it needs 2.6 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Jan v3 4B Base Instruct GGUF?
At Q4_K_M, Jan v3 4B Base Instruct GGUF can reach ~1104 tok/s on AMD Instinct MI300X. On NVIDIA GeForce RTX 4090: ~248 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: AMD Instinct MI300X → 5300 ÷ 2.6 × 0.55 = ~1104 tok/s
Estimated speed at Q4_K_M (2.6 GB)
AMD Instinct MI300X~1104 tok/sNVIDIA GeForce RTX 4090~248 tok/sNVIDIA H100 SXM~825 tok/sAMD Instinct MI250X~683 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Jan v3 4B Base Instruct GGUF?
At Q4_K_M, the download is about 2.40 GB. The full-precision Q8_0 version is 4.00 GB. The smallest option (Q3_K_S) is 1.75 GB.