BuilderBuilder
Versioned benchmark collection

Space Flight v2

Harness v2 · 16 measured runs across 1 target. Each entry is a single recorded attempt; repeat-run reliability is not established.

Inspect the run ledgers← All collections
Astronaut approaching a landed spacecraft in the Space Flight v2 collection
Astronaut approaching a landed spacecraft in the Space Flight v2 collection
16Measured runs
1Frozen targets
16Model families
16Configurations
Collection scope

Recorded evidence, without a synthetic score.

Runs share the same frozen target and prompt within each test, with disclosed provider-specific stacks. Costs are recorded estimates as labelled in each ledger, not subscription invoices.

Targets

  • Space Flight Stage 2 — Landfall

Model coverage

  • DeepSeek V4 Flash · Max
  • DeepSeek V4 Pro · Max
  • Claude Fable 5 · Max
  • Claude Fable 5.1 · Max
  • GLM 5.3 Flash · Max
  • GPT-5.6 Luna · Max
  • GPT-5.6 Sol · Ultra
  • GPT-6 Astra · Max
  • GPT-6 Sol · Max
  • Grok 4.6 · xhigh
  • Grok 4.7 · xhigh
  • Kimi K3 · Max
  • MiMo V2.6 Pro · High
  • Claude Opus 5 · Max
  • Claude Opus 5.5 · Unreported
  • Qwen3.8 Max · xhigh

Evidence available

  • 16 measured run ledgers
  • Four public presentation recordings
  • Cost, token, workflow timing, and artifact receipts
  • Prompt SHA-256 d725486150092800139ff8a37732ddac62948b4241156511a4e16248b9aeb27b
  • Recorded prompt path benchmarks/prompts/space-flight-game-threejs-stage-2-landfall.v1.md

Still missing

  • Repeat-run reliability evidence
  • Standardized gameplay and physical-input verification
  • Comparable local FPS measurements
  • 2 related records are listed separately and are not attached to this prompt.
Public receipts

16 measured run ledgers

Open any row for its recorded workflow cost, timing, token usage, stack disclosure, artifact integrity, and explicit evidence gaps.

01DeepSeek V4 FlashMax reasoning · Space Flight Stage 2 — Landfallpartial_token_timing_source_build_ledger · not_run · Prompt version v1$0.27 API-equivalent usage accounting, not an itemized subscription cash chargerecorded estimate4h 36m 23.0s wall-clockEnd-to-end workflowOpen 02DeepSeek V4 ProMax reasoning · Space Flight Stage 2 — Landfallpartial_generation_metrics_build_and_failed_runtime_acceptance_ledger · FAIL_ACCEPTANCE · Prompt version v1$0.50 API-equivalent usage accounting, not an itemized subscription cash chargerecorded estimate2h 17m 5.4s wall-clockEnd-to-end workflowOpen 03Claude Fable 5Max reasoning · Space Flight Stage 2 — Landfallpartial_token_timing_source_build_ledger · Prompt version v1$280.14 Claude API list-price equivalent, not an itemized Claude Code chargerecorded estimate2h 05m 20.4s wall-clockEnd-to-end workflowOpen 04Claude Fable 5.1Max reasoning · Space Flight Stage 2 — Landfallpartial_token_timing_source_build_and_two_cycle_browser_replay_ledger · PASS_WITH_LIMITATIONS · Prompt version v1$186.71 Official Anthropic API-list-price equivalent, not an itemized Claude Code subscription charge.recorded estimate2h 07m 34.4s wall-clockEnd-to-end workflowOpen 05GLM 5.3 FlashMax reasoning · Space Flight Stage 2 — Landfallpartial_multi_turn_token_timing_and_artifact_ledger · Prompt version v1$19.53 UNVERIFIED reference scenario: GLM-5.2/GLM-5.1 official rates applied to GLM-5.3 token counts; not a verified GLM-5.3 charge or subscription marginal-cash invoicerecorded estimate3h 36m 29.2s wall-clockEnd-to-end workflowOpen 06GPT-5.6 LunaMax reasoning · Space Flight Stage 2 — Landfallpartial_token_timing_production_build_and_supplied_acceptance_evidence_ledger · Prompt version v1$0.69 Official OpenAI standard API-list-price equivalent; not an itemized ChatGPT Pro/Codex subscription chargerecorded estimate1h 28m 30.7s wall-clockEnd-to-end workflowOpen 07GPT-5.6 SolUltra reasoning · Space Flight Stage 2 — Landfallpartial_token_timing_source_build_static_acceptance_ledger · Prompt version v1$20.84 Official OpenAI standard API-list-price equivalent; not an itemized Codex Pro subscription chargerecorded estimate1h 29m 57.7s wall-clockEnd-to-end workflowOpen 08GPT-6 AstraMax reasoning · Space Flight Stage 2 — Landfallpartial_cumulative_stage1_and_landfall_root_metrics_clean_build_and_model_supplied_stage2_evidence · Prompt version v1$188.63 Official OpenAI Standard API-list-price equivalent for the cumulative Stage 1 plus Landfall root-task receipt; not a Stage-2-only cost or an itemized Codex subscription chargerecorded estimateNot recordedEnd-to-end workflowOpen 09GPT-6 SolMax reasoning · Space Flight Stage 2 — Landfallcomplete_by_model_internal_verdict_with_qualified_visual_and_perf_scope · Prompt version v1$7.02 API-equivalent list price, not a marginal subscription cash chargerecorded estimate4707.4s wall-clockEnd-to-end workflowOpen 10Grok 4.6xhigh reasoning · Space Flight Stage 2 — Landfallpartial_one_shot_token_timing_production_build_and_scripted_acceptance_ledger · production_build_and_scripted_two_cycle_check_passed · Prompt version v1$4.32 Provider-recorded usage estimate, not a subscription cash-charge invoicerecorded estimate26m 36.958s wall-clockEnd-to-end workflowOpen 11Grok 4.7xhigh reasoning · Space Flight Stage 2 — Landfallcomplete_model_internal_judge_passed · PASS — model-internal full round trip and limited independent browser check · Prompt version v1$5.83 Provider-recorded usage estimate, not an itemized subscription cash chargerecorded estimateNot recordedEnd-to-end workflowOpen 12Kimi K3Max reasoning · Space Flight Stage 2 — Landfallpartial_token_timing_source_build_ledger · Prompt version v1$15.58 API-equivalent usage accounting, not an itemized subscription cash chargerecorded estimate2h 23m 18.1s wall-clockEnd-to-end workflowOpen 13MiMo V2.6 ProHigh reasoning · Space Flight Stage 2 — Landfallcomplete_by_model_internal_verdict_with_operator_caveats · Prompt version v1$1.01 Provider-recorded usage estimate, not an itemized subscription cash chargerecorded estimate6527.5s wall-clockEnd-to-end workflowOpen 14Claude Opus 5Max reasoning · Space Flight Stage 2 — Landfallpartial_token_timing_source_build_ledger · Prompt version v1$133.83 Claude API list-price equivalent, not an itemized Claude Code chargerecorded estimate2h 31m 33.5s wall-clockEnd-to-end workflowOpen 15Claude Opus 5.5Unreported reasoning · Space Flight Stage 2 — Landfallcomplete_by_model_internal_verdict_with_input_only_verification_gap · Prompt version v1$33.62 API-equivalent list price, not a marginal subscription cash chargerecorded estimateNot recordedEnd-to-end workflowOpen 16Qwen3.8 Maxxhigh reasoning · Space Flight Stage 2 — Landfallartifact_runtime_ledger_partial_metrics_no_blind_evaluation · PASS · Prompt version v1¥55.93 Independently recomputed first-party API-list-price equivalent for the measured token mix; NOT a billed amount. This run used an Alibaba Model Studio token-plan subscription, under which usage is not itemized as cash.recorded estimate1h 50m 56.0s wall-clockEnd-to-end workflowOpen
Prompt provenance

Exact prompt receipts

A path is shown only when the published record includes the prompt hash that identifies it. Campaign evidence and projections keep the hash gap visible.

space-flight-game-threejs-stage-2-landfall

SHA-256 d725486150092800139ff8a37732ddac62948b4241156511a4e16248b9aeb27b

Path benchmarks/prompts/space-flight-game-threejs-stage-2-landfall.v1.md

This recorded path has no released Builder prompt download.

Separate records

Related records with unknown prompt provenance

These published records stay separate from the prompt receipts above. Unknown provenance is why they are listed here.

Limits

Evidence and material gaps

  • No released Builder project, harness package, or skill download.
  • The recorded prompt path is not a released download.
  • Repeat-run reliability evidence is not established.