perrylink's picture
docs: correct E1 escalate numbers per 2026-10-08 verification
2411df9 verified
|
Raw History Blame
5.86 kB

Gallery — application scenarios

Animated demos of Phocinae-Largha-150M-v1 in typical agent workflows. All 27 demos: 800×450, loop forever, first frame = conclusion card with key numbers, last frame = one-line takeaway. All in-frame numbers match BENCHMARKS.md and MODEL_CARD.md.

Highlights

−54% LLM calls (82.8% at τ=0.5), kept-subset accuracy 0.886 (vs 0.797 local-only)

Command gate: rm -rf blocked in 18.6 ms

Tool routing: 12/12 tool selection

Three lines to run: pip install → serve → one decision

Bilingual decisions: typed acc en 0.797 / zh 0.789

Real-time event triage inside a 30 fps frame budget

Index

Approval safety

# scenario demo key numbers
S01 destructive command gate S01_rmrf_gate.gif p(deny)=0.96 · 18.6 ms
S02 curl-pipe-shell gate S02_curl_pipe_gate.gif p(allow)=0.21 → DENY · 18.6 ms
S03 batch command triage S03_chmod_batch.gif 8 commands: 6 allow / 1 ask / 1 deny
S04 tri-state approval S04_tristate_gate.gif thresholds 0.30/0.65; git push 0.91 / sudo restart 0.44 / rm -rf /etc 0.05
S05 replay battery S05_replay_battery.gif 17 commands (9 benign / 8 dangerous); 0 false allows · 2 false rejects

Routing & cost savings

# scenario demo key numbers
S06 escalate savings S06_escalate_savings.gif acc 0.797→0.886 (kept subset) · −54% LLM calls (82.8% at τ=0.5) (τ=0.6)
S07 tool routing S07_tool_routing.gif p(python)=0.87 · JevBench tool_selection 12/12
S08 Chinese zero-escalate S08_zh_zero_escalate.gif Largha 0.83 vs Kimi K3 0.72 (200 translated decisions)
S09 context screening S09_context_screen.gif 30 chunks: keep 19 / drop 11 · CPU 8-thread batch 8–21 decisions/s
S10 model routing S10_model_routing.gif simple → local 18.6 ms; complex → escalate (τ=0.6)

Real-time & fun

# scenario demo key numbers
S11 microbatch 60 fps S11_microbatch_60fps.gif 18.6 ms → micro-batch-4 amortized 6.2 ms · p(RIGHT)=0.92
S12 snake with a 144M brain S12_snake.gif 18.6 ms/step, 20×20 grid
S13 real-time event triage S13_event_triage.gif 30 fps frame gate · P(3)=0.81 red alert · 18.6 ms/event

Office & documents

# scenario demo key numbers
S14 document triage S14_doc_triage.gif 8 files labeled locally · p(contract)=0.88 · zero cloud tokens
S15 expense pre-approval S15_expense_preapprove.gif taxi ¥3,800 no receipt · p(approve)=0.07 → REJECT · gray band → human
S16 skill routing S16_skill_routing.gif one sentence → contract-parsing skill · p=0.94
S17 quality gate S17_quality_gate.gif draft report p(rewrite)=0.61 → flagged back · cheap first screen, not final review
S18 invoice verification S18_invoice_verify.gif ¥12,600×13% = ¥1,638.79 tax (matches) · p(ok)=0.95

Workflow & engineering

# scenario demo key numbers
S19 step verification S19_step_verify.gif expected 200 OK vs actual HTTP 500 · p(pass)=0.04 → FAIL, stop chain
S20 output screening S20_output_screen.gif p(usable)=0.35 block saves LLM tokens · first coarse screen only
S21 content gate S21_content_gate.gif allow / review / block (0.30/0.65 dual thresholds) · gray band → human
S22 option-order invariance S22_flip_invariance.gif flip rate 0.0300 — same verdict after option shuffle
S23 bilingual decision S23_bilingual.gif typed acc en 0.797 / zh 0.789 · zh = translated cases
S24 quickstart in three lines S24_quickstart.gif pip install → serve → POST · hardware tiers: 4 GB no-GPU 1.5–1.7 s / 3060-class 30–60 ms / 4090-class 18.6 ms

Head-to-head & reliability

# scenario demo key numbers
G25 local vs API race G25_race_local_vs_api.gif 18.6 ms local vs 1.51 s API round-trip (81×)
G26 token savings G26_token_savings.gif 100 decisions: 54 local / 46 escalated · −54% LLM calls (82.8% at τ=0.5)
G27 fail-closed gate G27_failclosed.gif server unreachable → DENY by default, no silent pass

Honest notes

  • All demos are synthesized (fictional scenario data) for capability illustration — not real user data
  • Thresholds shown in S04/S15/S21 are demo values; the shipped guard's thresholds are documented in deployment.md
  • In-frame numbers match BENCHMARKS.md; limitations (e.g. JevBench 0.5108, gate not passed) are disclosed in MODEL_CARD.md

Checksums: see SHA256SUMS.