Clarify Frozen80 score differences and link author homepage

Place direct model download links beside performance results

Place direct model download links beside performance results

Place direct model download links beside performance results

Update canonical model series and public download statistics

Update canonical model series and public download statistics

Update canonical model series and public download statistics

Fix family download metadata and counted download flows (#2)

Render technical report equations as vector mathematics

Highlight compute-aware architecture and organize technical details

Highlight compute-aware architecture and organize technical details

Condense network context into chart and report footnotes

Restore four-model latency chart with updated TypeSafe percentiles

Document candidate-aware projection and update TypeSafe HTTP reference

Synchronize technical report, final HTTP charts and dataset descriptions

Correct reported 4B HTTP P50 to18ms

Add 4B latency values calculated from reported percentile ratios

Clarify high-effort HTTP metrics and single-GPU chart labels

Verify vLLM usage on GPU and update latency and effort charts

Add evaluation dataset link and vLLM serving example

Publish complete APUS-OpenJev-v1 models and Technical Report v1.1