Versioned benchmark collection

Campfire Interactive Scene v1

Harness v1 · 23 measured runs across 1 target. Each entry is a single recorded attempt; repeat-run reliability is not established.

Campfire benchmark output in the Campfire Interactive Scene v1 collection
Campfire benchmark output in the Campfire Interactive Scene v1 collection
23Measured runs
1Frozen targets
18Model families
21Configurations
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

  • Campfire under the stars

Model coverage

  • DeepSeek V4 Flash · Vision exp
  • DeepSeek V4.1 Flash · Highest
  • Claude Fable 5 · Max
  • Claude Fable 5 · Medium
  • Claude Fable 5.1 · Max
  • Gemini 3.8 Flash · Highest
  • GLM 5.3 Flash · Max
  • GPT-5.6 Luna · Max
  • GPT-5.6 Luna · xhigh
  • GPT-5.6 Sol · Ultra
  • GPT-5.6 Sol · xhigh
  • GPT-5.6 Terra · Ultra
  • GPT-6 Astra · Max
  • GPT-6 Sol · Max
  • Grok 4.6 · xhigh
  • Kimi K3 · Max
  • Claude Opus 5 · Max
  • Qwen3.8 Flash · Unreported
  • Qwen3.8 Max Preview · Unreported
  • Qwen3.8 Max · xhigh
  • Claude Sonnet 5 · Max

Evidence available

  • 23 measured run ledgers
  • Public browser builds
  • Cost, token, and workflow timing
  • Prompt SHA-256 81bb5ab559c254d69ba83368aef23c89c4b10f0692e232b8b8f400b6e9ed8cd3
  • Recorded prompt path benchmarks/prompts/campfire-threejs.v1.md

Still missing

  • Browser/hardware environment
  • Local FPS for several runs
  • Final capture metadata
  • Minimum blind-vote sample
  • 5 ledgers in this scope do not record a prompt hash. They remain in the historical scope and are not proof of the frozen prompt.
  • 7 related records are listed separately and are not attached to this prompt.
Public receipts

23 measured run ledgers

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

01DeepSeek V4 FlashVision exp reasoning · Campfire under the starsPASS · Prompt version unreported$0.08 Provider-recorded OpenRouter costrecorded estimateNot recordedEnd-to-end workflowOpen 02DeepSeek V4.1 FlashHighest reasoning · Campfire under the starscomplete_token_timing_artifact_and_validation_ledger · pass · Prompt version unreported$0.53 Provider-recorded usage estimate, not an itemized subscription cash chargerecorded estimate4227.4s wall-clockEnd-to-end workflowOpen 03Claude Fable 5Max reasoning · Campfire under the starspartial_token_timing_and_source_ledger · Prompt version v1$11.17 API-equivalent estimate, not a subscription invoicerecorded estimate12:48.9 wall-clockEnd-to-end workflowOpen 04Claude Fable 5Medium reasoning · Campfire under the starspartial_token_timing_and_source_ledger · Prompt version unreported · Prompt hash unavailablePrompt hash was not recorded on this historical ledger. It stays in the historical collection scope and is not proof of the frozen prompt.$5.12 Anthropic first-party Claude API, standard global pricingrecorded estimate4:24.6 wall-clockEnd-to-end workflowOpen 05Claude Fable 5.1Max reasoning · Campfire under the starspartial_artifact_static_serve_and_rounded_usage_ledger · Prompt version v1$18.89 Provider-reported session cost; not independently recomputable from the rounded display.recorded estimate59m (rounded provider display) wall-clockEnd-to-end workflowOpen 06Gemini 3.8 FlashHighest reasoning · Campfire under the starscomplete_token_timing_artifact_and_validation_ledger · pass · Prompt version unreported$4.44 Provider-recorded usage estimate, not an itemized subscription cash chargerecorded estimate4287.5s wall-clockEnd-to-end workflowOpen 07GLM 5.3 FlashMax reasoning · Campfire under the starsPASS · Prompt version unreported$0.02 Provider-recorded OpenRouter costrecorded estimate7m 21.718s wall-clockEnd-to-end workflowOpen 08GLM 5.3 FlashMax reasoning · Campfire under the starspartial_one_shot_runtime_visual_failure · FAIL_VISUAL_VALIDITY · Prompt version v1$0.11 OpenCode-recorded API-equivalent usage accounting, not an itemized subscription marginal-cash invoicerecorded estimate2m 40.772s wall-clockEnd-to-end workflowOpen 09GPT-5.6 LunaMax reasoning · Campfire under the starsPARTIAL · Prompt version unreportedNot recordedrecorded estimateNot recordedEnd-to-end workflowOpen 10GPT-5.6 LunaMax reasoning · Campfire under the starspartial_token_timing_and_artifact_ledger · Prompt version v1$0.06 API-equivalent estimate, not a subscription invoicerecorded estimate17m 11.6s wall-clockEnd-to-end workflowOpen 11GPT-5.6 Lunaxhigh reasoning · Campfire under the starspartial_token_timing_and_source_ledger · Prompt version v1 · Prompt hash unavailablePrompt hash was not recorded on this historical ledger. It stays in the historical collection scope and is not proof of the frozen prompt.$0.56 API-equivalent estimate, not a subscription invoicerecorded estimate8:40.2 wall-clockEnd-to-end workflowOpen 12GPT-5.6 SolUltra reasoning · Campfire under the starspartial_token_timing_and_source_ledger · Prompt version v1 · Prompt hash unavailablePrompt hash was not recorded on this historical ledger. It stays in the historical collection scope and is not proof of the frozen prompt.$6.86 API-equivalent estimate, not a subscription invoicerecorded estimate23:08.4 wall-clockEnd-to-end workflowOpen 13GPT-5.6 Solxhigh reasoning · Campfire under the starspartial_token_timing_and_source_ledger · Prompt version v1 · Prompt hash unavailablePrompt hash was not recorded on this historical ledger. It stays in the historical collection scope and is not proof of the frozen prompt.$2.26 API-equivalent estimate, not a subscription invoicerecorded estimate28:52.6 wall-clockEnd-to-end workflowOpen 14GPT-5.6 TerraUltra reasoning · Campfire under the starspartial_token_timing_and_source_ledger · Prompt version v1 · Prompt hash unavailablePrompt hash was not recorded on this historical ledger. It stays in the historical collection scope and is not proof of the frozen prompt.$1.86 API-equivalent estimate, not a subscription invoicerecorded estimate24:28.8 wall-clockEnd-to-end workflowOpen 15GPT-6 AstraMax reasoning · Campfire under the starspartial_qualified_root_task_snapshot_with_build_and_live_load · Prompt version v1$15.49 Standard API-equivalent estimate from a qualified root-task snapshot, not a subscription invoicerecorded estimateNot recordedEnd-to-end workflowOpen 16GPT-6 SolMax reasoning · Campfire under the starscomplete_token_timing_artifact_and_validation_ledger · Prompt version unreported$3.43 API-equivalent list price, not a marginal subscription cash chargerecorded estimateNot recordedEnd-to-end workflowOpen 17Grok 4.6xhigh reasoning · Campfire under the starspartial_one_shot_metrics_source_build_and_browser_smoke_ledger · Prompt version v1$2.37 Provider-recorded usage estimate, not a subscription cash-charge invoicerecorded estimate13m 13.687s wall-clockEnd-to-end workflowOpen 18Kimi K3Max reasoning · Campfire under the starspartial_token_timing_and_source_ledger · Prompt version v1$0.45 API-equivalent usage accounting, not an itemized subscription cash chargerecorded estimate14:19.3 wall-clockEnd-to-end workflowOpen 19Claude Opus 5Max reasoning · Campfire under the starspartial_token_timing_artifact_build_ledger · Prompt version v1$27.45 First-party Claude API list-price equivalent, not a subscription invoicerecorded estimate39m 37.2s wall-clockEnd-to-end workflowOpen 20Qwen3.8 FlashUnreported reasoning · Campfire under the starsPENDING_JUDGE · Prompt version unreported$0.19 Provider-recorded OpenRouter costrecorded estimate55m 21.924s wall-clockEnd-to-end workflowOpen 21Qwen3.8 Max PreviewUnreported reasoning · Campfire under the starspartial_token_timing_and_source_ledger · Prompt version v1$0.30 Third-party NanoGPT API-list-price equivalent for the measured token mix; not an Alibaba Token Plan cash chargerecorded estimate6m 0.9s wall-clockEnd-to-end workflowOpen 22Qwen3.8 Maxxhigh reasoning · Campfire under the starsartifact_model_runtime_metrics_ledger_missing_fps_and_blind_evaluation · PASS · Prompt version v1¥20.79 First-party API-list-price-equivalent upper bound, not a token-plan cash charge (upper bound)recorded upper bound15m 33.8s wall-clockEnd-to-end workflowOpen 23Claude Sonnet 5Max reasoning · Campfire under the starspartial_token_timing_and_artifact_ledger · Prompt version v1$19.22 First-party Claude API list-price equivalent, not a subscription invoicerecorded estimate49m 25.0s 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.

campfire-threejs

SHA-256 81bb5ab559c254d69ba83368aef23c89c4b10f0692e232b8b8f400b6e9ed8cd3

Path benchmarks/prompts/campfire-threejs.v1.md

Open the released prompt pack
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

  • Repeat-run reliability evidence is not established.