MiniCPM V 4.7 35B A3B DSpark — Hardware Requirements & GPU Compatibility
Chat
MiniCPM V 4.7 35B A3B DSpark is a 35B-parameter open language model from OpenBMB in the MiniCPM family. It supports a context window of up to 262,144 tokens. At Q4_K_M it needs about 21.32 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 16 GB (Q2_K · 64K)Best on 24 GB (Q4_K_M · ~326 tok/s · 256K)Macs from 32 GB (~44 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run MiniCPM V 4.7 35B A3B DSpark
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve openbmb/MiniCPM-V-4.7-35B-A3B-DSpark --max-model-len 16384vLLM repo with its own quantisation
Download: Weights on Hugging Face
Other ways
each opens step-by-stepRun in cloud
Fits on RTX A6000 (48 GB) (26 GB headroom) · Q4_K_M
- Generation speed
- ~254 tok/s
- generation speed
- Cost per 1M output tokens
- $0.36
- per 1M output tokens
How Much VRAM Does MiniCPM V 4.7 35B A3B DSpark Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 15.2 GB | 15.5 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 256K context · ~442 tok/s | |
| 17.4 GB | 17.7 GB | 3-bit medium — 3-bit medium quantization | Fits: 256K context · ~392 tok/s | |
| 21.3 GB | 21.6 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 256K context · ~326 tok/s | |
| 25.3 GB | 25.6 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Does not fit: needs 25.4 GB | |
| 29.2 GB | 29.5 GB | 6-bit, very good — 6-bit quantization, very good quality | Does not fit: needs 29.4 GB | |
| 35.3 GB | 35.6 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Does not fit: needs 35.5 GB | |
| 70.3 GB | 70.6 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Does not fit: needs 70.5 GB |
Selected Q4_K_M
4-bit medium quantization21.5GBat 16K context
- Weights
- 21.0 GB
- KV cache
- 0.2 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 2.5 GB free · ~326 tok/s
≥50 tok/s needs ≥142 GB/s · RTX 5090 ~542, RTX 3090 Ti ~326
Which Hardware Can Run MiniCPM V 4.7 35B A3B DSpark?
Q4_K_M · 21.3 GBMiniCPM V 4.7 35B A3B DSpark (Q4_K_M) requires 21.3 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 28+ GB is recommended. Using the full 262K context window can add up to 2.7 GB, bringing total usage to 24.0 GB. 7 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 103 devices with unified memory can run MiniCPM V 4.7 35B A3B DSpark, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Studio M1 Max (32 GB).
20 GB
Q4_K_M · best fit Q3_K_M (17.5 GB)
1 GPU
RX 7900 XT~244Simulate output on RX 7900 XT32–40 GB
fits · 10.5 GB free · up to 256K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~542Simulate output on RTX 5090M5 Max36 GB~150Simulate output on Mac Studio M5 Max (36 GB)M532 GB~56Simulate output on MacBook Air 13" M5 (32 GB)Orin~71Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~47Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Alternatives to MiniCPM V 4.7 35B A3B DSpark
MiniCPM family · 16 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Qwen3.6 35B A3BAlibaba | 107.1 | 22 GB | ~151 | same size |
| Llm Jp 4.1 33B Thinkingllm-jp | — | 22.5 GB | ~35 | same size |
| Agnes 3.0 FlashAgnes-AI | — | 22.1 GB | ~35 | same size |
| Ornith 1.5 35B A3Bornith-ai | — | 22 GB | ~151 | same size |
| Agents A1InternScience | — | 21.5 GB | ~171 | same size |
| Intern S2 PreviewInternLM | — | 73.2 GB | —needs > 24 GB | same size |
Specifications
- Publisher
- OpenBMB
- Family
- MiniCPM
- Parameters
- 35B
- Active parameters
- 3B per token
- Architecture
- Qwen3DSparkModel
- Context Length
- 262,144 tokens
- Vocabulary Size
- 248,144
- Release Date
- 2026-10-08
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does MiniCPM V 4.7 35B A3B DSpark need?
MiniCPM V 4.7 35B A3B DSpark requires 21.3 GB of VRAM at Q4_K_M, or 70.3 GB at BF16. Full 262K context adds up to 2.7 GB (24.0 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 35B × 4.8 bits ÷ 8 = 21 GB
KV Cache + Overhead ≈ 0.3 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 ≈ 3 GB (at full 262K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 4090 run MiniCPM V 4.7 35B A3B DSpark?
Yes, at Q4_K_M (21.3 GB) or lower. Higher quantizations like Q5_K_M (25.3 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
What's the best quantization for MiniCPM V 4.7 35B A3B DSpark?
For MiniCPM V 4.7 35B A3B DSpark, Q4_K_M (21.3 GB) offers the best balance of quality and VRAM usage. Q5_K_M (25.3 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 15.2 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run MiniCPM V 4.7 35B A3B DSpark on a Mac?
Yes, but only at lower quantizations. The smallest Mac that can run MiniCPM V 4.7 35B A3B DSpark is Mac Mini M2 (24 GB) at Q2_K; 107 of the 156 Macs we list can run it at some quantization. For Q4_K_M (21.3 GB) you need a Mac with more unified memory.
Can I run MiniCPM V 4.7 35B A3B DSpark locally?
Yes — MiniCPM V 4.7 35B A3B DSpark can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is MiniCPM V 4.7 35B A3B DSpark?
At Q4_K_M, MiniCPM V 4.7 35B A3B DSpark can reach ~724 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~326 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = 1000 ÷ (active GB ÷ (bandwidth GB/s × efficiency) × 1000 + layers × routing ms)
Mixture-of-Experts: just the active experts get read for each token, along with a fixed routing cost per layer.
Example: NVIDIA B200 → 1.8 GB active ÷ (8000 × 0.65) = 0.3514 ms, plus 5 layers × 0.055 ms = 0.2750 ms, so 1000 ÷ 0.6264 ms = ~1596 tok/s
Estimated speed at Q4_K_M (21.3 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of MiniCPM V 4.7 35B A3B DSpark?
At Q4_K_M, the download is about 21.00 GB. The full-precision BF16 version is 70.00 GB. The smallest option (Q2_K) is 14.88 GB.
Which GPUs can run MiniCPM V 4.7 35B A3B DSpark?
7 consumer GPUs can run MiniCPM V 4.7 35B A3B DSpark at Q4_K_M. Top options include NVIDIA GeForce RTX 5090, AMD Radeon RX 7900 XTX, NVIDIA GeForce RTX 3090. 1 GPU have plenty of headroom for comfortable inference.
Which devices can run MiniCPM V 4.7 35B A3B DSpark?
103 devices with unified memory can run MiniCPM V 4.7 35B A3B DSpark at Q4_K_M (21.3 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB), Beelink GTR9 Pro (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.