Cloud baseline

GLM 5.3 Flash — 0.0000 on Forge 1.0

z-ai/glm-5.3-flash · scorer forge-1.0

GLM 5.3 Flash, single model via OpenRouterz-ai/glm-5.3-flash · 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
0.0000
Benchmark
Forge 1.0
Model build
23m 15s

Recorded inference cost

$0.38

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 ($0.30673218, incomplete) was replaced; the benchmark score and run are unchanged.

Token usage

Tokens: 7.8M in · 59k out · 7.6M cached (exact, OpenRouter records)

7,835,065 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: 25,775 (not added to the total).

  • z-ai/glm-5.3-flash-20260826

    In 7.8M · Out 59k · Cached 7.6M · Reasoning 25.8k · 151 calls

Tier breakdown

L · Lint and bundles0.00
K · Platform currency0.00
T · Event pipeline0.00
R · Reconcile0.00
S · Storage0.00
B · Resolvers and permissions0.00
U · UI function0.00
V · Visual0.00
A · Rovo0.00
E · Excellence0.00

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 × 0.0000 core + 0.12 × 0.00 gate × 0.00 excellence) × 0.2160 critical = 0.0000

Critical defects compound a multiplier on the whole score (pre-severity 0.0000):

l_deployable 0.00 → ×0.60r_backfill_complete 0.00 → ×0.60u_widget_loads 0.00 → ×0.60

l_bundles_load: no additional penalty (vacuous:precondition: >= 1 function declared); crash — the app does not run.

t_no_double_count: no additional penalty (vacuous:precondition: >= 1 ledger row); wrong numbers — a change counted twice.

b_no_permission_leak: no additional penalty (vacuous:precondition: >= 1 person-facing change list returned or rendered); data leak — hidden issues shown to a person.

b_comment_exactly_once: no additional penalty (vacuous:precondition: >= 1 comment POST); duplicate side effect on a customer's Jira.

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 (0 of 18 met here), then pays out at the excellence tier's own measured mean. Core 0.0000 + excellence 0.0000.

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

Earned score 0.000 · Cap 0.499 · Final score 0.000

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

  • deployable: maximum 0.499l_deployablel_bundles_load
  • working ledger: maximum 0.699u_widget_loadst_event_rowsr_backfill_completes_storage_scopes_entity_index_used
  • current platform, complete surfaces: maximum 0.799k_dashboard_widgetk_widget_edit_bridgek_rovo_skillu_widget_edit_configu_ledger_tablea_action_resultv_theme_tokensv_dark_mode
  • production robustness: maximum 0.799t_no_double_countt_out_of_ordert_retry_after_honouredr_heal_droppedr_rate_limitr_paginationb_no_permission_leakb_comment_exactly_onceb_realtime_payload_cleanu_llm_explainu_widget_livev_csp_cleanv_console_cleanu_ledger_sorts_index_orderb_hidden_counta_action_permissionsl_scopesk_llm_model_current

Forge's own linter reports no errors (warnings cost points), every manifest function bundles and loads, packages resolve at the kit's pins, the manifest rules the linter misses, and only the scopes the app's calls need.

l deployable

0.00

manifest.yml does not parse: All mapping items must start at the same column at line 86, column 1: - claude resources: ^

l bundles load

0.00

vacuous — precondition unmet: >= 1 function declared

l lint warnings

0.00

vacuous — precondition unmet: >= 1 function loads

l real packages

0.00

vacuous — precondition unmet: >= 1 bundle imports @forge/*

l manifest rules

0.00

vacuous — precondition unmet: >= 1 resource and >= 1 function

l scopes

0.00

vacuous — precondition unmet: >= 1 product or KVS call observed

Graded browser recording

No recording: no app surface ever rendered.

v_theme_tokens, v_dark_mode, v_csp_clean, v_console_clean: vacuous — precondition unmet: a surface rendered app content

Run details

Model
z-ai/glm-5.3-flash
Engine events
0
Repair rounds
0
Started
Oct 6, 2026, 02:10 AM
Finished
Oct 6, 2026, 02:35 AM

How the benchmark is scored