perrylink commited on
Commit
f7bea7a
·
verified ·
1 Parent(s): ae78820

assets: new demos/charts + cleaned docs

Browse files
Files changed (1) hide show
  1. docs/gallery/README.md +10 -1
docs/gallery/README.md CHANGED
@@ -1,6 +1,6 @@
1
  # Gallery — application scenarios
2
 
3
- Animated demos of Phocinae-Largha-150M-v1 in typical agent workflows. All 24 demos: 800×450, loop forever, first frame = conclusion card with key numbers, last frame = one-line takeaway. All in-frame numbers match [BENCHMARKS.md](../../BENCHMARKS.md) and [MODEL_CARD.md](../../MODEL_CARD.md).
4
 
5
  ## Highlights
6
 
@@ -70,6 +70,15 @@ Animated demos of Phocinae-Largha-150M-v1 in typical agent workflows. All 24 dem
70
  | S23 | bilingual decision | [S23_bilingual.gif](./S23_bilingual.gif) | typed acc en 0.797 / zh 0.789 · zh = translated cases |
71
  | S24 | quickstart in three lines | [S24_quickstart.gif](./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 |
72
 
 
 
 
 
 
 
 
 
 
73
  ## Honest notes
74
 
75
  - All demos are **synthesized** (fictional scenario data) for capability illustration — not real user data
 
1
  # Gallery — application scenarios
2
 
3
+ 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](../../BENCHMARKS.md) and [MODEL_CARD.md](../../MODEL_CARD.md).
4
 
5
  ## Highlights
6
 
 
70
  | S23 | bilingual decision | [S23_bilingual.gif](./S23_bilingual.gif) | typed acc en 0.797 / zh 0.789 · zh = translated cases |
71
  | S24 | quickstart in three lines | [S24_quickstart.gif](./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 |
72
 
73
+
74
+ ### Head-to-head & reliability
75
+
76
+ | # | scenario | demo | key numbers |
77
+ |---|---|---|---|
78
+ | G25 | local vs API race | [G25_race_local_vs_api.gif](./G25_race_local_vs_api.gif) | 18.6 ms local vs 1.51 s API round-trip (81×) |
79
+ | G26 | token savings | [G26_token_savings.gif](./G26_token_savings.gif) | 100 decisions: 82 local / 18 escalated · −82% LLM calls |
80
+ | G27 | fail-closed gate | [G27_failclosed.gif](./G27_failclosed.gif) | server unreachable → DENY by default, no silent pass |
81
+
82
  ## Honest notes
83
 
84
  - All demos are **synthesized** (fictional scenario data) for capability illustration — not real user data