hf_mirrors/MiniCPM-V-4.7-35B-A3B-DSpark
Model Introduction
Files & versions
Analysis

MiniCPM-V-4.7-35B-A3B-DSpark

MiniCPM-V-4.7-35B-A3B-DSpark is a DSpark draft checkpoint trained for exact pairing with MiniCPM-V-4.7-35B-A3B and its tokenizer.

The llama.cpp version please see: MiniCPM-V-4.7-35B-A3B-DSpark-gguf

Model Specification

PropertyValue
Target modelMiniCPM-V-4.7-35B-A3B
Draft layers5
Draft parameters389,223,169
Draft tokens per forward pass15
Target hidden-state layers[3, 11, 19, 27, 35]
Checkpoint precisionBF16

Training Recipe

Training responses were generated by MiniCPM-V-4.7-35B-A3B from a mixture of Vision & Text prompts.

SettingValue
Training sequences2M
Epochs6
OptimizerAdamW
ObjectiveCE + L1 + confidence loss

Evaluation

Acceptance length is defined as the total number of completion tokens divided by the total number of speculative verification steps. Evaluation uses natural EOS termination and max_new_tokens=8192.

DomainAccept Length
Vision
MMMU5.098
MathVista6.126
CharXiv_reasoning_val5.479
OCRBench5.407
Text
Math6.687
Code7.387
General3.501

Usage with vLLM

vllm serve \
  --model openbmb/MiniCPM-V-4.7-35B-A3B \
  --served-model-name MiniCPM-V-4.7-35B-A3B \
  --tensor-parallel-size 2 \
  --max-model-len 40960 \
  --max-num-batched-tokens 40960 \
  --max-num-seqs 1 \
  --attention-backend FLASH_ATTN \
  --mm-encoder-attn-backend FLASH_ATTN \
  --mm-processor-kwargs '{"downsample_mode":"4x"}' \
  --trust-request-mm-kwargs \
  --disable-custom-all-reduce \
  --compilation-config '{"pass_config":{"fuse_allreduce_rms":false}}' \
  --speculative-config '{"method":"dspark","model":"openbmb/MiniCPM-V-4.7-35B-A3B-DSpark","num_speculative_tokens":15,"enable_adaptive_verification":false}' \
  --default-chat-template-kwargs '{"enable_thinking":false}'

Limitations and Disclaimer

This model has no autonomous intent or legal personhood; its outputs are text generated from statistical patterns and may be inaccurate, biased, or offensive, and may be manipulated by carefully crafted prompts ("jailbreaks") into producing unintended content. Its responses on sensitive topics such as politics, health, finance, and law are not reviewed by experts and should not be treated as professional advice.

This model is provided "AS IS", without warranty of any kind, express or implied, and the developers are not liable for any damages arising from its use. Users must employ the model only for lawful, compliant, and ethical purposes, configure their own safeguards, and label AI-generated content where required; deliberate jailbreaking, injection attacks, or inducing harmful output is prohibited, and any such testing is at the user's own risk.