MiniCPM5-2B-DSpark
openbmbapache-2.0text

MiniCPM5-2B-DSpark

Updated Nov 14, 2026
Parameters
300M
Context
131,072
License
apache-2.0
Updated
Nov 14, 2026

Will it run on your hardware?

Pick your GPU memory - see which quantizations fit, and the cheapest card for the rest

Runs on your 24 GB - best at FP16
1 of 1 quantizations fit MiniCPM5-2B-DSpark with real runtime overhead.
FP16
1 GB
fits tight too big

Need an exact figure for your context length? Use the VRAM calculator.

Run it locally

Copy-paste - running in under a minute

vLLMOpenAI-compatible API
vllm serve openbmb/MiniCPM5-2B-DSpark

New to this? Start with Ollama · serve to many users with vLLM.

Deep dive

Notes, sources, and the full write-up

MiniCPM5-2B-DSpark is a 0.3B-parameter apache-2.0 model from openbmb. At Q4_K_M it needs roughly 0 GB of VRAM, placing it in the CPU / ≤4 GB hardware tier.

PopularityLink to this section

MiniCPM5-2B-DSpark has 1,001,270 downloads in the last month on HuggingFace and 45 likes.

From the community

What users are saying

“T denotes the sampling temperature. T=0 uses greedy decoding; T=1.0 uses sampling at temperature 1.0. Both use the same draft checkpoint and preserve the target's output distribution under the respective decoding settings.”

“In Evaluation, what is the difference between T=0 and T=1, and why is there such a large discrepancy in acceptance rate?”

Frequently asked

Quick answers to common questions

How much VRAM does MiniCPM5-2B-DSpark need?

MiniCPM5-2B-DSpark with 300M parameters needs approximately 0 GB at Q4_K_M quantization. Use our VRAM calculator for an exact estimate.

Is MiniCPM5-2B-DSpark better than other openbmb models?

MiniCPM5-2B-DSpark has 300M parameters with 131,072 context - a strong choice for long-context, edge, general-chat.

What license is MiniCPM5-2B-DSpark under?

MiniCPM5-2B-DSpark is released under the apache-2.0 license, making it suitable for most commercial and personal projects.

What hardware runs MiniCPM5-2B-DSpark well?

With 300M parameters, MiniCPM5-2B-DSpark is small enough for most consumer GPUs, Apple Silicon Macs, and many CPUs. Check our hardware directory for specific recommendations.

What is the best quantization for MiniCPM5-2B-DSpark?

Q4_K_M is the recommended sweet spot - ~98% of FP16 quality at ~27% of the size. Q5_K_M (~0 GB) is an option if you have spare VRAM. Use our VRAM calculator to compare.

How long can MiniCPM5-2B-DSpark's context window handle?

MiniCPM5-2B-DSpark supports a 131,072-token context window - enough for very long documents, codebases, or multi-turn conversations. Real-world usable context may vary by implementation.

What models compete with MiniCPM5-2B-DSpark?

MiniCPM5-2B-DSpark competes with other models in its class. Browse our model directory for comparisons, benchmarks, and community reviews to find the best fit.

Nearby options

Similar models and compatible hardware by spec