Cloud baseline

Hy4 preview — 0.6890 on Forge 1.0

tencent/hy4-preview · scorer forge-1.0

Hy4 preview, single model via OpenRoutertencent/hy4-preview · single agentOct 6, 2026
Historic benchmarkForge 1.0 · not comparable with Forge 2.0

Forge 1.0 was frozen on 9 October 2026 and accepts no new runs; Forge 2.0 is its successor. The original results contained scorer faults. Corrections made from saved evidence on 15 October are dated on the affected run pages, with both original and corrected scores and the reason for the change. They reuse the original builds and observations; no benchmark or model was rerun, and the original inference charges remain. Uncorrected results retain known limitations: the top five original scores, 0.945 to 0.977, were too close to reliably rank the strongest models; two hourly-job tests missed Jira reads performed by a queue worker; and the old linter accepted an entity index with two range attributes that Forge refuses. Those remaining faults are not silently treated as fixed. Consult each run's correction note when comparing historical results. Forge 2.0 has separate results and a different task.

Overall
0.6890
Benchmark
Forge 1.0
Model build
37m 52s

Recorded inference cost

$1.36

From the saved OpenRouter generation billing summary (151 recorded requests). Whether auxiliary model calls were included was not retained. Scoring reuses the saved build; this amount describes the original inference spend.

  • Billing recovered 2026-10-18 from all 151 of this build's own generation ids with OpenRouter's generation records (151 billed, native token reconciliation passed). The earlier saved summary ($1.348705235, incomplete) was replaced; the benchmark score and run are unchanged.

Token usage

Tokens: 20.2M in · 120k out · 19.9M cached (exact, OpenRouter records)

20,277,044 tokens over 151 model calls: every call of the run, the model's own and the harness's helper calls, counted from OpenRouter records. Reasoning tokens as the provider reports them: 65,450 (not added to the total).

  • tencent/hy4-preview-20260827

    In 20.2M · Out 120k · Cached 19.9M · Reasoning 65.5k · 151 calls

Tier breakdown

L · Lint and bundles1.00
K · Platform currency0.90
T · Event pipeline0.75
R · Reconcile0.9041
S · Storage1.00
B · Resolvers and permissions0.7222
U · UI function0.6384
V · Visual0.8588
A · Rovo1.00
E · Excellence0.544

Scoring detail

How this score was built

Per-test scorer output, as posted by the app. Expand a section to inspect its tests.

Correctness combines the lowest score for each test across the scoring fixtures; economy measurements use their mean. The combined result can therefore be lower than every individual fixture’s final score. Critical defects reduce the weighted result before any score cap applies.

Earned credit before admission

(0.88 × 0.8341 core + 0.12 × 0.82 gate × 0.67 excellence) × 1.0000 critical = 0.7993

The core (88% of the total) is the weighted mean of the nine measured tiers. The last 12% is the excellence slice: it unlocks in proportion to the perfection conditions below (13 of 18 met here), then pays out at the excellence tier's own measured mean. Core 0.7340 + excellence 0.0653.

t_event_rows 0.00r_heal_dropped 1.00u_widget_numbers 0.75u_ledger_table 0.61b_comment_adf_as_user 0.33k_consumer_shape 1.00k_current_apis 1.00k_dashboard_widget 1.00k_entity_declared 1.00k_llm_model_current 0.00k_manifest_semantics 1.00k_rovo_mcp 1.00k_rovo_skill 1.00k_runtime_risks 1.00k_widget_edit_bridge 1.00v_console_clean 1.00v_csp_clean 1.00l_lint_warnings 1.00

Earned score 0.799 · Cap 0.699 · Final score 0.689

The final score is the lower of the earned score and the cap. Meeting a cap's conditions adds no points.

  • working ledger: maximum 0.699t_event_rows
  • current platform, complete surfaces: maximum 0.799u_ledger_tablev_theme_tokensv_dark_mode
  • production robustness: maximum 0.799t_out_of_orderr_paginationb_comment_exactly_onceu_llm_explainu_widget_liveu_ledger_sortk_llm_model_current

Few Jira requests in the backfill and per event, few round trips before each surface paints, and a scheduled run with nothing new writes nothing — paid only in proportion to the excellence gate.

e reconcile economy

1.00

mean over 3 scoring sites: backfill 15 Jira calls / optimum 15 (incl. 2 forced 429 repeat(s)) = 1.00x

e event economy

0.00

mean over 3 scoring sites: event rows inexact — economy is not credited on unfinished work

e ui round trips

1.00

mean over 3 scoring sites: round trips before first paint: {'widget-view': 0, 'widget-edit': 1, 'sprint-action': 1}

Run notes and corrections

Scorer correction (2026-10-15): 0.4796 → 0.6890 from the same saved build. The original scorer applied a duplicate-comment penalty when an action failed its comment-count or success-flag check, even though no duplicate was recorded. All retained fixtures recorded at most one comment per action. The failed comment-action checks and original admission ceiling remain in the score. No model, benchmark, candidate application or browser probe was rerun. Original inference charges are unchanged. Legacy observation files lacked an embedded original observation hash; this correction uses their retained native build, fixture identities and original functional-grade matches.

Graded browser recording

Loading duration…

Watch the full graded browser recording at its original speed: the dashboard widget, the sprint action and the not-started sprint, as the scorer drove them on the run's own seed. The test results and screenshots on this page provide the wider evidence. Pause or scrub to inspect a view.

Ready to play0:00 / —:—
Open recording file

Drag the timeline to seek. Keyboard: arrow keys move 5 seconds; Home and End jump to the start and end.

Full graded browser recording: dashboard widget (no config, edit + Save, light and dark, 380 and 1180 px, a second instance), sprint action modal (sort, router, select, comment post through the 429 retry, double click, forbidden post, close) and the not-started sprint

Screenshots

Browser screenshots of the built app, taken during grading. Each caption names the view.

1Every captured surface
2Widget edit view
3Widget before configuration
4Sprint not started
5Sprint action · dark

Run details

Model
tencent/hy4-preview
Engine events
0
Repair rounds
0
Started
Oct 6, 2026, 04:51 PM
Finished
Oct 6, 2026, 05:37 PM
Re-scored
Oct 15, 2026, 07:14 AM (same build, corrected scorer)

How the benchmark is scored