Model comparison

DeepSeek-V4-Provsall-MiniLM-L6-v2

Decision Card metrics — latency, license, lineage overlap, benchmarks, and repo health from the live graph.

Decision metrics

AttributeDeepSeek-V4-Proall-MiniLM-L6-v2
Licenseunknownunknown
ParametersModel size
Context window
Architecturetext-generationsentence-similarity
Hardware class
Latency
$ / 1M tokens
Verified TTFT
Repo health (RIS)
Graph trust8888
Lineage1 paper3 papers
Graph score5.1K5K

Decision cards

DeepSeek-V4-Pro
Licenseunknown
Parameters
Context
Architecturetext-generation
Lifecycle
BenchmarksPending verified runs
Repo healthNo linked repo
EconomicsEconomics pending verified runs
all-MiniLM-L6-v2
Licenseunknown
Parameters
Context
Architecturesentence-similarity
Lifecycle
Base: nreimers/MiniLM-L6-H384-uncased
BenchmarksPending verified runs
Repo healthNo linked repo
EconomicsEconomics pending verified runs

DeepSeek-V4-Pro

Hugging Face model with 5135 likes. Tags: transformers, safetensors, deepseek_v4, text-generation, conversational, arxiv:2606.19348, license:mit, eval-results, endpoints_compatible, 8-bit

all-MiniLM-L6-v2

Hugging Face model with 5026 likes. Tags: sentence-transformers, pytorch, tf, rust, onnx, safetensors, openvino, bert, feature-extraction, sentence-similarity