kazeia/kazeia-android/app
Kazeia Team 3d435f9cdd LLM: trim system prompt to drop ~27 prefill tokens (-1.3s TTFT)
The verbose 55-token system prompt was the cheapest TTFT win on the
kv-only path (52 ms per prefill token). Compacting it to 25 tokens while
keeping the three load-bearing constraints — Kazeia identity, French only,
short replies, /no_think — measurably improved end-to-end latency.

Validated 'Bonjour, comment vas-tu ?' on tablet:
  Before: prompt_tokens=80, TTFT=4202ms, total=5716ms
  After:  prompt_tokens=53, TTFT=2865ms, total=4034ms (-1.3s, -32% TTFT)

Reply quality preserved: "Bonjour ! Je vais bien, merci. Comment vas-tu ?"

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-14 12:16:11 +02:00
..
src/main LLM: trim system prompt to drop ~27 prefill tokens (-1.3s TTFT) 2026-04-14 12:16:11 +02:00
build.gradle.kts TTS tremor investigation: identify cross-arch numerical floor, gate diag flags 2026-04-13 00:15:14 +02:00
proguard-rules.pro Initial commit: Kazeia TTS pipeline on NPU via ExecuTorch 2026-04-09 08:42:11 +02:00