MiniCPM5 1B SFT — Hardware Requirements & GPU Compatibility
Chat
MiniCPM5 1B SFT is a 1.1B-parameter open language model from openbmb. It supports a context window of up to 131,072 tokens. At Q4_K_M it needs about 0.99 GB of VRAM — see which GPUs and Macs can run it below.
8.5K downloads
25 likesSpecifications
- Publisher
- openbmb
- Parameters
- 1.1B
- Architecture
- LlamaForCausalLM
- Context Length
- 131,072 tokens
- Vocabulary Size
- 130,560
- Release Date
- 2026-05-25
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does MiniCPM5 1B SFT Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q4_K_M | 4.80 | 1.0 GB | 3.4 GB | 0.65 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_M | 5.70 | 1.1 GB | 3.5 GB | 0.77 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 1.2 GB | 3.6 GB | 0.89 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 1.4 GB | 3.8 GB | 1.08 GB | 8-bit quantization, near-lossless |
Which GPUs Can Run MiniCPM5 1B SFT?
Q4_K_M · 1.0 GBMiniCPM5 1B SFT (Q4_K_M) requires 1.0 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 2+ GB is recommended. Using the full 131K context window can add up to 2.4 GB, bringing total usage to 3.4 GB. 35 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.
Runs great
— Plenty of headroomWhich Devices Can Run MiniCPM5 1B SFT?
Q4_K_M · 1.0 GB33 devices with unified memory can run MiniCPM5 1B SFT, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
Runs great
— Plenty of headroomRelated Models
Frequently Asked Questions
- How much VRAM does MiniCPM5 1B SFT need?
MiniCPM5 1B SFT requires 1.0 GB of VRAM at Q4_K_M, or 1.4 GB at Q8_0. Full 131K context adds up to 2.4 GB (3.4 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 1.1B × 4.8 bits ÷ 8 = 0.6 GB
KV Cache + Overhead ≈ 0.4 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 2.8 GB (at full 131K context)
VRAM usage by quantization
Q4_K_M1.0 GBQ4_K_M + full context3.4 GB- What's the best quantization for MiniCPM5 1B SFT?
For MiniCPM5 1B SFT, Q4_K_M (1.0 GB) offers the best balance of quality and VRAM usage. Q5_K_M (1.1 GB) provides better quality if you have the VRAM.
VRAM requirement by quantization
Q4_K_M ★1.0 GBQ5_K_M1.1 GBQ6_K1.2 GBQ8_01.4 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run MiniCPM5 1B SFT on a Mac?
MiniCPM5 1B SFT requires at least 1.0 GB at Q4_K_M, 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 MiniCPM5 1B SFT locally?
Yes — MiniCPM5 1B SFT can run locally on consumer hardware. At Q4_K_M quantization it needs 1.0 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is MiniCPM5 1B SFT?
At Q4_K_M, MiniCPM5 1B SFT can reach ~2944 tok/s on AMD Instinct MI300X. On NVIDIA GeForce RTX 4090: ~662 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 ÷ 1.0 × 0.55 = ~2944 tok/s
Estimated speed at Q4_K_M (1.0 GB)
~2944 tok/s~662 tok/s~2201 tok/s~1820 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of MiniCPM5 1B SFT?
At Q4_K_M, the download is about 0.65 GB. The full-precision Q8_0 version is 1.08 GB.
- Which GPUs can run MiniCPM5 1B SFT?
35 consumer GPUs can run MiniCPM5 1B SFT at Q4_K_M (1.0 GB). Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 35 GPUs have plenty of headroom for comfortable inference.
- Which devices can run MiniCPM5 1B SFT?
33 devices with unified memory can run MiniCPM5 1B SFT at Q4_K_M (1.0 GB), including Mac Mini M4 (16 GB), Mac Mini M4 (32 GB), Mac Mini M4 Pro (24 GB), Mac Mini M4 Pro (48 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.