Loading the gallery…
Runs lost to infrastructure (not results)
Infrastructure failures (not model results)
Attempts that ended because of the harness launcher, the connection or a provider's quota (a free route's rate limit), not because of the model: 118 so far, 89 of them quota refusals. They are not results: they are left out of the result list, the counts, the per-model totals and averages and the head-to-heads on Compare. Where a cell was run again, the later run is its result.
| Model and harness | Prompt | Attempt | How it ended | Wall time | Cost | Date | Note | Result |
|---|---|---|---|---|---|---|---|---|
| DeepSeek V4.1 FlashOpenCode | MacBook Pro 16" realistic 3D render (Playcode benchmark) | 1 | Infrastructure failure | 30m 32s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-09-23 | WhyInfrastructure failure, not a model result: the prompt was never delivered, because the harness was not ready when the runner sent it, and the run was stopped after 30 minutes without progress. The cell was re-run. | the cell's result |
| GLM-5.3Claude Code | Pelican riding a bicycle (control) | 1 | Infrastructure failure | 6m 26s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyInfrastructure failure, not a model result: the runner addressed the Claude Code tmux session by a name containing "." (glm-5.3), which tmux reads as a window.pane separator, so the prompt was never delivered and 0 tokens reached the model ($0). Runner bug fixed 05:06Z; cell re-run. | the cell's result |
| GLM-5.3Claude Code | Pelican riding a bicycle (control) | 2 | Infrastructure failure | 6m 26s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyInfrastructure failure, not a model result: the runner addressed the Claude Code tmux session by a name containing "." (glm-5.3), which tmux reads as a window.pane separator, so the prompt was never delivered and 0 tokens reached the model ($0). Runner bug fixed 05:06Z; cell re-run. | the cell's result |
| Grok 4.7Grok Build | Animated pelican riding a bicycle (SMIL/CSS) | 1 | Infrastructure failure | 21 s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyInfrastructure failure, not a model result: the provider stopped serving requests mid-run. No tokens were used and nothing was produced. | Pending — queued, not run yet |
| GLM-5.3Claude Code | DualSense controller, max fidelity | 1 | Infrastructure failure | 9m 33s | $0.002 estimated | 2026-09-24 | WhyInfra failure, not a model DNF: after 8.5 min the Fireworks stream ended with 'API Error: Stream ended without receiving any events' (1.2k input / 14 output tokens, 0 tool calls, no artifact). Re-queued as a re-run. | the cell's result |
| Kimi K3Kimi Code CLI | California brown pelican riding a bicycle (detailed spec) | 1 | Infrastructure failure | 17m 23s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyInfra failure, not a model DNF: after 17 min of reasoning (about 95k characters) the model wrote one line ("I'll create this SVG step by step:") and the response stream closed with no tool call and no final usage record, so no brown_pelican.svg was written. Every other completed step on this endpoint across the 7 Kimi Code runs reported usage; this one reported 0 input / 0 output, which marks a truncated stream. Kimi Code's own local count put the context at about 24.6k tokens. Not re-run yet. | the cell's result |
| Gemini 3.8 FlashCline CLI | BMW M4 Competition, side view (flat vector) | 1 | Infrastructure failure | 50 s | $0 reported by Cline CLI | 2026-09-24 | WhyAfter 50 s the Cline gateway (Vercel -> Google) answered HTTP 400 'Request contains an invalid argument' (one Vertex attempt inside it gave 503); Cline ended the run. Upstream gateway failure (the same error on attempt 4 shows the chain: Cline's Google key over quota 429 -> Vertex 503 -> final fallback 400); not a model failure. | the cell's result |
| Gemini 3.8 FlashCline CLI | BMW M4 Competition, side view (flat vector) | 3 | Infrastructure failure | 21m 53s | $0 estimated | 2026-09-24 | WhyAfter 103 iterations (22 min, no bmw_m4.svg written yet) the Cline gateway (Vercel -> Google) answered HTTP 400 'Request contains an invalid argument'; Cline ended the run. Upstream gateway failure (the same error on attempt 4 shows the chain: Cline's Google key over quota 429 -> Vertex 503 -> final fallback 400); not a model failure. (Record salvaged by hand: the runner's ledger step crashed on this attempt; fixed.) | the cell's result |
| Gemini 3.8 FlashCline CLI | BMW M4 Competition, side view (flat vector) | 4 | Quota refusal | 48 s | $0 reported by Cline CLI | 2026-09-24 | WhyCline free tier answered a rate limit / quota error ('quota'); upstream of the harness, not a model failure. [Reclassified 2026-09-24T20:05Z from INFRASTRUCTURE: a 429 quota refusal is QUOTA and does not count against the row.] | the cell's result |
| Gemini 3.8 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 1 | Infrastructure failure | 2m 39s | $0 reported by Cline CLI | 2026-09-24 | WhyCline's gateway (Vercel -> Google/Vertex) failed the request mid-run (provider attempts: Google 400 'invalid argument' / Vertex 503 'temporarily unavailable'); Cline ended the run. Upstream of the harness, not a model failure. | the cell's result |
| Gemini 3.8 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 2 | Infrastructure failure | 48 s | $0 reported by Cline CLI | 2026-09-24 | WhyCline's gateway (Vercel -> Google/Vertex) failed the request mid-run (provider attempts: Google 400 'invalid argument' / Vertex 503 'temporarily unavailable'); Cline ended the run. Upstream of the harness, not a model failure. | the cell's result |
| Gemini 3.8 FlashCline CLI | California brown pelican riding a bicycle (detailed spec) | 1 | Infrastructure failure | 1m 12s | $0 reported by Cline CLI | 2026-09-24 | WhyCline's gateway (Vercel -> Google/Vertex) failed the request mid-run (provider attempts: Google 400 'invalid argument' / Vertex 503 'temporarily unavailable'); Cline ended the run. Upstream of the harness, not a model failure. | the cell's result |
| Gemini 3.8 FlashCline CLI | Xbox 360 controller | 1 | Infrastructure failure | 44 s | $0 reported by Cline CLI | 2026-09-24 | WhyCline's gateway (Vercel -> Google/Vertex) failed the request mid-run (provider attempts: Google 400 'invalid argument' / Vertex 503 'temporarily unavailable'); Cline ended the run. Upstream of the harness, not a model failure. | the cell's result |
| Gemini 3.8 FlashCline CLI | Lighthouse on a headland at night | 1 | Infrastructure failure | 23 s | $0 reported by Cline CLI | 2026-09-24 | WhyCline's gateway (Vercel -> Google/Vertex) failed the request mid-run (provider attempts: Google 400 'invalid argument' / Vertex 503 'temporarily unavailable'); Cline ended the run. Upstream of the harness, not a model failure. | the cell's result |
| Gemini 3.8 FlashCline CLI | MacBook Pro 16" realistic 3D render (Playcode benchmark) | 1 | Infrastructure failure | 4m 24s | $0 reported by Cline CLI | 2026-09-24 | WhyCline's gateway (Vercel -> Google/Vertex) failed the request mid-run (provider attempts: Google 400 'invalid argument' / Vertex 503 'temporarily unavailable'); Cline ended the run. Upstream of the harness, not a model failure. | the cell's result |
| Gemini 3.8 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 3 | Infrastructure failure | 4m 45s | $0 reported by Cline CLI | 2026-09-24 | WhyCline's gateway (Vercel -> Google/Vertex) failed the request mid-run (Vertex 503 'temporarily unavailable', then Google 400 'invalid argument'); Cline ended the run (exit 1). A pelican_animated.svg existed when the run was cut (kept for reference, not a finished result). Not a model failure. First filed as MODEL/dnf by the runner, reclassified. | the cell's result |
| Gemini 3.8 FlashCline CLI | Xbox 360 controller | 2 | Infrastructure failure | 3m 37s | $0 reported by Cline CLI | 2026-09-24 | WhyCline's gateway (Vercel -> Google/Vertex) failed the request mid-run (provider attempts: Google 400 'invalid argument' / Vertex 503 'temporarily unavailable'); Cline ended the run. Upstream of the harness, not a model failure. A xbox360.svg existed when the run was cut (kept for reference, not a finished result). | the cell's result |
| Gemini 3.8 FlashCline CLI | Fjord scene (Three.js) | 1 | Quota refusal | 7.6 s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyCline's daily free-usage limit for cline-free/gemini-3.8-flash was reached before the run could start ('Daily free model limit reached … Try again in 21h 53m', ~17:47Z 2026-09-25). Not a model failure; first filed as MODEL/dnf by the runner, reclassified. | no result yet |
| Gemini 3.8 FlashCline CLI | DualSense controller, max fidelity | 2 | Quota refusal | 8.6 s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyCline's daily free-usage limit for cline-free/gemini-3.8-flash was reached before the run could start ('Daily free model limit reached … Try again in 21h 53m', ~17:47Z 2026-09-25). Not a model failure; first filed as MODEL/dnf by the runner, reclassified. | the cell's result |
| Gemini 3.8 FlashCline CLI | Isometric SVG town with sprite sheet | 2 | Quota refusal | 8.0 s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyCline's daily free-usage limit for cline-free/gemini-3.8-flash was reached before the run could start ('Daily free model limit reached … Try again in 21h 53m', ~17:47Z 2026-09-25). Not a model failure; first filed as MODEL/dnf by the runner, reclassified. | the cell's result |
| Gemini 3.8 FlashCline CLI | Animated short (HTML/JS) | 2 | Quota refusal | 7.8 s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyCline's daily free-usage limit for cline-free/gemini-3.8-flash was reached before the run could start ('Daily free model limit reached … Try again in 21h 53m', ~17:47Z 2026-09-25). Not a model failure; first filed as MODEL/dnf by the runner, reclassified. | no result yet |
| DeepSeek V4.1 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 1 | Quota refusal | 7m 18s | $0 reported by Cline CLI | 2026-09-24 | WhyCline's daily free-usage limit for cline-free/deepseek-v4.1-flash was reached ('daily free model limit'); resets in about 23h 22m (~2026-09-25T19:55:07Z). Quota, not a model failure; does not count against the row's attempts. A pelican_animated.svg existed when the run was cut (kept for reference, not a finished result). | the cell's result |
| Gemini 3.8 FlashCline CLI | MacBook Pro 16" realistic 3D render (Playcode benchmark) | 3 | Quota refusal | 7.7 s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyCline's daily free-usage limit for cline-free/gemini-3.8-flash was reached ('daily free model limit'); resets in about 20h 26m (~2026-09-25T17:47:50Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| DeepSeek V4.1 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 2 | Quota refusal | 7.7 s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyCline's daily free-usage limit for cline-free/deepseek-v4.1-flash was reached ('daily free model limit'); resets in about 21h 42m (~2026-09-25T19:55:04Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| MiMo V2.6 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 1 | Quota refusal | 43m 21s | $0 reported by Cline CLI | 2026-09-24 | WhyCline's daily free-usage limit for cline-free/mimo-v2.6-flash was reached ('daily free model limit'); resets in about 21h 41m (~2026-09-25T20:37:28Z). Quota, not a model failure; does not count against the row's attempts. A pelican_animated.svg existed when the run was cut (kept for reference, not a finished result). | the cell's result |
| Gemini 3.8 FlashCline CLI | MacBook Pro 16" realistic 3D render (Playcode benchmark) | 4 | Quota refusal | 9.0 s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyCline's daily free-usage limit for cline-free/gemini-3.8-flash was reached ('daily free model limit'); resets in about 18h 50m (~2026-09-25T17:46:54Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| Muse Spark 1.3Cline CLI | Pelican riding a bicycle (control) | 1 | Quota refusal | 9.7 s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyCline's free route answered a quota / rate-limit refusal ('429'); model held until 2026-09-24T23:02:08Z. Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| Muse Spark 1.3Cline CLI | Pelican riding a bicycle (control) | 2 | Quota refusal | 49 s | $0 reported by Cline CLI | 2026-09-24 | WhyCline's free route answered a quota / rate-limit refusal ('429'); model held until 2026-09-24T23:09:42Z. Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| Muse Spark 1.3Cline CLI | Pelican riding a bicycle (control) | 3 | Quota refusal | 14 s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyCline's free route answered a quota / rate-limit refusal ('status 429'); model held until 2026-09-24T23:15:12Z. Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| DeepSeek V4.1 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 3 | Quota refusal | 7.8 s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyCline's daily free-usage limit for cline-free/deepseek-v4.1-flash was reached ('daily free model limit'); resets in about 20h 38m (~2026-09-25T19:55:03Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| Muse Spark 1.3Cline CLI | Pelican riding a bicycle (control) | 4 | Quota refusal | 1m 04s | $0 reported by Cline CLI | 2026-09-24 | WhyCline's free route answered a quota / rate-limit refusal ('status 429'); model held until 2026-09-24T23:23:10Z. Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| Muse Spark 1.3Cline CLI | California brown pelican riding a bicycle (detailed spec) | 1 | Quota refusal | 19 s | $0 reported by Cline CLI | 2026-09-24 | WhyCline's free route answered a quota / rate-limit refusal ('status 429'); model held until 2026-09-24T23:53:30Z. Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| Muse Spark 1.3Cline CLI | California brown pelican riding a bicycle (detailed spec) | 2 | Quota refusal | 8.3 s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyCline's free route answered a quota / rate-limit refusal ('status 429'); model held until 2026-09-24T23:58:48Z. Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| MiMo V2.6 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 2 | Quota refusal | 8.5 s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyCline's daily free-usage limit for cline-free/mimo-v2.6-flash was reached ('daily free model limit'); resets in about 20h 41m (~2026-09-25T20:37:38Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| Gemini 3.8 FlashCline CLI | MacBook Pro 16" realistic 3D render (Playcode benchmark) | 5 | Quota refusal | 8.5 s | not recorded: the run's usage was never recorded | 2026-09-24 | WhyCline's daily free-usage limit for cline-free/gemini-3.8-flash was reached ('daily free model limit'); resets in about 17h 50m (~2026-09-25T17:47:03Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| Muse Spark 1.3Cline CLI | California brown pelican riding a bicycle (detailed spec) | 3 | Quota refusal | 20 s | $0 reported by Cline CLI | 2026-09-24 | WhyCline's free route answered a quota / rate-limit refusal ('status 429'); model held until 2026-09-25T00:04:09Z. Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| Muse Spark 1.3Cline CLI | Lighthouse on a headland at night | 1 | Quota refusal | 3m 27s | $0 reported by Cline CLI | 2026-09-25 | WhyCline's free route answered a quota / rate-limit refusal ('status 429'); model held until 2026-09-25T00:20:27Z. Quota, not a model failure; does not count against the row's attempts. A lighthouse.svg existed when the run was cut (kept for reference, not a finished result). | the cell's result |
| DeepSeek V4.1 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 4 | Quota refusal | 8.5 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/deepseek-v4.1-flash was reached ('daily free model limit'); resets in about 19h 38m (~2026-09-25T19:55:13Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| Muse Spark 1.3Cline CLI | Lighthouse on a headland at night | 2 | Quota refusal | 3m 35s | $0 reported by Cline CLI | 2026-09-25 | WhyCline's free route answered a quota / rate-limit refusal ('429'); model held until 2026-09-25T00:29:03Z. Quota, not a model failure; does not count against the row's attempts. A lighthouse.svg existed when the run was cut (kept for reference, not a finished result). | the cell's result |
| Gemini 3.8 FlashCline CLI | MacBook Pro 16" realistic 3D render (Playcode benchmark) | 6 | Quota refusal | 7.8 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/gemini-3.8-flash was reached ('daily free model limit'); resets in about 16h 28m (~2026-09-25T17:47:41Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| DeepSeek V4.1 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 5 | Quota refusal | 7.9 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/deepseek-v4.1-flash was reached ('daily free model limit'); resets in about 18h 36m (~2026-09-25T19:55:51Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| MiMo V2.6 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 3 | Quota refusal | 7.5 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/mimo-v2.6-flash was reached ('daily free model limit'); resets in about 19h 17m (~2026-09-25T20:37:01Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| Claude Opus 5.5none (raw API) | Fjord scene (Three.js) | 3 | Infrastructure failure | 15 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyThe connection upstream of the harness dropped the run (API Error: 502 <!DOCTYPE html> <!--[if lt IE 7]> <html class="no-js ie6 oldie" lang="en-US"> <![endif]--> <!--[if IE 7]> <html class="no-js ie7 oldie" lang="en…); not a model failure, not our config. | the cell's result |
| Gemini 3.8 FlashCline CLI | MacBook Pro 16" realistic 3D render (Playcode benchmark) | 7 | Quota refusal | 7.6 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/gemini-3.8-flash was reached ('daily free model limit'); resets in about 15h 17m (~2026-09-25T17:47:20Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| DeepSeek V4.1 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 6 | Quota refusal | 7.6 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/deepseek-v4.1-flash was reached ('daily free model limit'); resets in about 17h 25m (~2026-09-25T19:55:30Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| MiMo V2.6 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 4 | Quota refusal | 7.7 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/mimo-v2.6-flash was reached ('daily free model limit'); resets in about 18h 7m (~2026-09-25T20:37:40Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| Muse Spark 1.3Cline CLI | Animated short (HTML/JS) | 1 | Quota refusal | 12m 07s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/muse-spark-1.3-contributor was reached ('daily free model limit'); resets in about 20h 16m (~2026-09-25T23:04:19Z). Quota, not a model failure; does not count against the row's attempts. A index.html existed when the run was cut (kept for reference, not a finished result). | no result yet |
| Gemini 3.8 FlashCline CLI | MacBook Pro 16" realistic 3D render (Playcode benchmark) | 8 | Quota refusal | 8.1 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/gemini-3.8-flash was reached ('daily free model limit'); resets in about 14h 17m (~2026-09-25T17:47:29Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| DeepSeek V4.1 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 7 | Quota refusal | 7.7 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/deepseek-v4.1-flash was reached ('daily free model limit'); resets in about 16h 25m (~2026-09-25T19:55:40Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| MiMo V2.6 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 5 | Quota refusal | 7.7 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/mimo-v2.6-flash was reached ('daily free model limit'); resets in about 17h 6m (~2026-09-25T20:36:51Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| Muse Spark 1.3Cline CLI | Animated short (HTML/JS) | 2 | Quota refusal | 7.5 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/muse-spark-1.3-contributor was reached ('daily free model limit'); resets in about 19h 16m (~2026-09-25T23:04:28Z). Quota, not a model failure; does not count against the row's attempts. | no result yet |
| Gemini 3.8 FlashCline CLI | MacBook Pro 16" realistic 3D render (Playcode benchmark) | 9 | Quota refusal | 8.1 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/gemini-3.8-flash was reached ('daily free model limit'); resets in about 13h 17m (~2026-09-25T17:47:38Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| DeepSeek V4.1 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 8 | Quota refusal | 10 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/deepseek-v4.1-flash was reached ('daily free model limit'); resets in about 15h 25m (~2026-09-25T19:55:52Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| MiMo V2.6 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 6 | Quota refusal | 7.8 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/mimo-v2.6-flash was reached ('daily free model limit'); resets in about 16h 6m (~2026-09-25T20:37:04Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| Muse Spark 1.3Cline CLI | Animated short (HTML/JS) | 3 | Quota refusal | 12 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/muse-spark-1.3-contributor was reached ('daily free model limit'); resets in about 18h 15m (~2026-09-25T23:03:42Z). Quota, not a model failure; does not count against the row's attempts. | no result yet |
| Gemini 3.8 FlashCline CLI | MacBook Pro 16" realistic 3D render (Playcode benchmark) | 10 | Quota refusal | 8.2 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/gemini-3.8-flash was reached ('daily free model limit'); resets in about 12h 17m (~2026-09-25T17:47:48Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| DeepSeek V4.1 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 9 | Quota refusal | 8.6 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/deepseek-v4.1-flash was reached ('daily free model limit'); resets in about 14h 24m (~2026-09-25T19:55:05Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| MiMo V2.6 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 7 | Quota refusal | 12 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/mimo-v2.6-flash was reached ('daily free model limit'); resets in about 15h 6m (~2026-09-25T20:37:21Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| Muse Spark 1.3Cline CLI | Animated short (HTML/JS) | 4 | Quota refusal | 8.0 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/muse-spark-1.3-contributor was reached ('daily free model limit'); resets in about 17h 15m (~2026-09-25T23:03:51Z). Quota, not a model failure; does not count against the row's attempts. | no result yet |
| Gemini 3.8 FlashCline CLI | MacBook Pro 16" realistic 3D render (Playcode benchmark) | 11 | Quota refusal | 8.4 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/gemini-3.8-flash was reached ('daily free model limit'); resets in about 11h 16m (~2026-09-25T17:46:58Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| DeepSeek V4.1 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 10 | Quota refusal | 8.5 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/deepseek-v4.1-flash was reached ('daily free model limit'); resets in about 13h 24m (~2026-09-25T19:55:14Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| MiMo V2.6 FlashCline CLI | Animated pelican riding a bicycle (SMIL/CSS) | 8 | Quota refusal | 8.1 s | not recorded: the run's usage was never recorded | 2026-09-25 | WhyCline's daily free-usage limit for cline-free/mimo-v2.6-flash was reached ('daily free model limit'); resets in about 14h 2m (~2026-09-25T20:37:32Z). Quota, not a model failure; does not count against the row's attempts. | the cell's result |
| GPT-6 AstraCodex CLI | Animated short (HTML/JS) | 1 | Infrastructure failure | 17m 43s | $3.16 estimated | 2026-09-25 | WhyThe prompt never reached the harness (turn 2 not seen in the session log after 3 Enter presses) (ours/infrastructure); not a model result. | the cell's result |
| GLM-5.3Claude Code | folding_road_threejs | 1 | Infrastructure failure | 4m 30s | not recorded: the run's usage was never recorded | 2026-09-26 | WhyThe connection upstream of the harness dropped the run (API Error: Stream ended without receiving any events); not a model failure, not our config. | no result yet |
| Claude Opus 5.5Claude Code | threejs_momijigari | 1 | Infrastructure failure | 20m 04s | $0.27 estimated | 2026-09-26 | WhyThe connection upstream of the harness dropped the run (API Error: Stream ended without receiving any events); not a model failure, not our config. | the cell's result |
| Kimi K3Kimi Code CLI | gpu_fluid_sim | 1 | Harness setup failure | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-09-26 | WhyOur harness/sandbox setup failed before the prompt was sent (harness exited before its ready screen); not a model result. | the cell's result |
| Claude Fable 5.1Claude Code | blender_lowpoly_dragon | 1 | Infrastructure failure | 47m 56s | $10.58 estimated | 2026-09-26 | WhyThe connection upstream of the harness dropped the run (API Error: 502 {"error":"relay upstream error"}. This is a server-side issue, usually temporary — try again in a moment. If it persists, check your inference g…); not a model failure, not our config. | no result yet |
| Claude Opus 5.5Claude Code | blender_lowpoly_dragon | 1 | Harness setup failure | 43m 22s | $6.87 estimated | 2026-09-26 | WhyOur harness configuration cut the run (API Error: Stream idle timeout - no chunks received); not a model failure. | no result yet |
| GPT-6 AstraCodex CLI | blender_lowpoly_dragon | 1 | Infrastructure failure | 43m 37s | $11.20 estimated | 2026-09-26 | WhyInfrastructure failure, not a model result: the run's last turn ended on HTTP 502 "relay upstream error" from the model relay, which went down with our coordinator machine at 15:49 UTC; the run was cut and wrote no render.png. The runner's finalizer had recorded it as "the model finished without writing render.png"; the transcript shows the 502, so it is listed here, not as a model DNF. | the cell's result |
| DeepSeek V4.1 Flashdsh | blender_lowpoly_dragon | 1 | Infrastructure failure | 43m 12s | $0.11 estimated | 2026-09-26 | WhyThe connection upstream of the harness dropped the run (DeepSeek API error (HTTP 502)); not a model failure, not our config. | the cell's result |
| Claude Fable 5.1OpenCode | blender_lowpoly_dragon | 1 | Infrastructure failure | 43m 44s | $9.96 reported by OpenCode | 2026-09-26 | WhyThe connection upstream of the harness dropped the run (APIError: Bad Gateway: relay upstream error); not a model failure, not our config. | no result yet |
| Claude Opus 5.5OpenCode | blender_lowpoly_dragon | 1 | Infrastructure failure | 43m 21s | $6.61 reported by OpenCode | 2026-09-26 | WhyThe connection upstream of the harness dropped the run (APIError: Bad Gateway: relay upstream error); not a model failure, not our config. | no result yet |
| Claude Fable 5.1Claude Code | Animated short (HTML/JS) | 1 | Infrastructure failure | 11m 20s | not recorded: the run's usage was never recorded | 2026-09-26 | WhyThe connection upstream of the harness dropped the run (API Error: Stream ended without receiving any events); not a model failure, not our config. | the cell's result |
| GLM-5.3Claude Code | threejs_dragon_ship | 1 | Infrastructure failure | 17m 02s | $0.20 estimated | 2026-09-26 | WhyThe connection upstream of the harness dropped the run (API Error: Stream ended without receiving any events); not a model failure, not our config. | no result yet |
| gemini-3.1-pro-previewnone (raw API) | threejs_artwork_style_pass | 1 | Quota refusal | 0.6 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-02 | WhyThe provider refused the request (HTTP 429: quota or account limit); no reply. | no result yet |
| Gemini 3.8 Flashnone (raw API) | threejs_artwork_style_pass | 1 | Quota refusal | 0.7 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-02 | WhyThe provider refused the request (HTTP 429: quota or account limit); no reply. | no result yet |
| Kimi K3none (raw API) | threejs_artwork_style_pass | 1 | Quota refusal | 0.5 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-02 | WhyThe provider refused the request (HTTP 429: quota or account limit); no reply. | no result yet |
| gemini-3.1-pro-previewnone (raw API) | threejs_mechanical_bee | 1 | Quota refusal | 0.6 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-02 | WhyThe provider refused the request (HTTP 429: quota or account limit); no reply. | no result yet |
| Gemini 3.8 Flashnone (raw API) | threejs_mechanical_bee | 1 | Quota refusal | 0.7 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-02 | WhyThe provider refused the request (HTTP 429: quota or account limit); no reply. | no result yet |
| Kimi K3none (raw API) | threejs_mechanical_bee | 1 | Quota refusal | 0.6 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-02 | WhyThe provider refused the request (HTTP 429: quota or account limit); no reply. | no result yet |
| Kimi K3none (raw API) | Synthwave hover-bike racer on a neon grid highway (Three.js) | 1 | Quota refusal | 0.3 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| Grok 4.7none (raw API) | Synthwave hover-bike racer on a neon grid highway (Three.js) | 1 | Quota refusal | 0.2 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 403: quota, [redacted] or account limit); no reply. | no result yet |
| Kimi K3none (raw API) | Floating pirate island diorama with a cliff cross-section (Three.js) | 1 | Quota refusal | 0.3 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| Grok 4.7none (raw API) | Floating pirate island diorama with a cliff cross-section (Three.js) | 1 | Quota refusal | 0.2 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 403: quota, [redacted] or account limit); no reply. | no result yet |
| gemini-3.1-pro-previewnone (raw API) | Synthwave hover-bike racer on a neon grid highway (Three.js) | 1 | Quota refusal | 0.5 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| gemini-3.5-flashnone (raw API) | Synthwave hover-bike racer on a neon grid highway (Three.js) | 1 | Quota refusal | 0.5 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| Gemini 3.8 Flashnone (raw API) | Synthwave hover-bike racer on a neon grid highway (Three.js) | 1 | Quota refusal | 0.6 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| gemini-3.1-pro-previewnone (raw API) | Floating pirate island diorama with a cliff cross-section (Three.js) | 1 | Quota refusal | 0.6 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| gemini-3.5-flashnone (raw API) | Floating pirate island diorama with a cliff cross-section (Three.js) | 1 | Quota refusal | 0.4 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| Gemini 3.8 Flashnone (raw API) | Floating pirate island diorama with a cliff cross-section (Three.js) | 1 | Quota refusal | 0.6 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| gemini-3.1-pro-previewnone (raw API) | Vintage biplane aerobatics over a grass airfield (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| gemini-3.5-flashnone (raw API) | Vintage biplane aerobatics over a grass airfield (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| Gemini 3.8 Flashnone (raw API) | Vintage biplane aerobatics over a grass airfield (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| Grok 4.7none (raw API) | Vintage biplane aerobatics over a grass airfield (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 403: quota, [redacted] or account limit); no reply. | no result yet |
| Kimi K3none (raw API) | Vintage biplane aerobatics over a grass airfield (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| gemini-3.1-pro-previewnone (raw API) | Retro arcade claw machine with pastel plush prizes (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| gemini-3.5-flashnone (raw API) | Retro arcade claw machine with pastel plush prizes (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| Gemini 3.8 Flashnone (raw API) | Retro arcade claw machine with pastel plush prizes (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| Grok 4.7none (raw API) | Retro arcade claw machine with pastel plush prizes (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 403: quota, [redacted] or account limit); no reply. | no result yet |
| Kimi K3none (raw API) | Retro arcade claw machine with pastel plush prizes (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-06 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| gemini-3.1-pro-previewnone (raw API) | Outer-space pinball table with planet bumpers (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-07 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| gemini-3.5-flashnone (raw API) | Outer-space pinball table with planet bumpers (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-07 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| Gemini 3.8 Flashnone (raw API) | Outer-space pinball table with planet bumpers (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-07 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| Grok 4.7none (raw API) | Outer-space pinball table with planet bumpers (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-07 | WhyThe provider refused the request (HTTP 403: quota, [redacted] or account limit); no reply. | no result yet |
| Kimi K3none (raw API) | Outer-space pinball table with planet bumpers (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-07 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| gemini-3.1-pro-previewnone (raw API) | Rube Goldberg breakfast machine on a kitchen countertop (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-07 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| gemini-3.5-flashnone (raw API) | Rube Goldberg breakfast machine on a kitchen countertop (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-07 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| Gemini 3.8 Flashnone (raw API) | Rube Goldberg breakfast machine on a kitchen countertop (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-07 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| Grok 4.7none (raw API) | Rube Goldberg breakfast machine on a kitchen countertop (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-07 | WhyThe provider refused the request (HTTP 403: quota, [redacted] or account limit); no reply. | no result yet |
| Kimi K3none (raw API) | Rube Goldberg breakfast machine on a kitchen countertop (Three.js) | 1 | Quota refusal | not recorded | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-07 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| gemini-3.5-flashnone (raw API) | Four-stroke engine cutaway: crank, piston and cam-driven valves (Three.js) | 1 | Quota refusal | 0.5 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-08 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| gemini-3.1-pro-previewnone (raw API) | Four-stroke engine cutaway: crank, piston and cam-driven valves (Three.js) | 1 | Quota refusal | 0.5 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-08 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| Grok 4.7none (raw API) | Four-stroke engine cutaway: crank, piston and cam-driven valves (Three.js) | 1 | Quota refusal | 0.6 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-08 | WhyThe provider refused the request (HTTP 403: quota, [redacted] or account limit); no reply. | no result yet |
| Kimi K3none (raw API) | Four-stroke engine cutaway: crank, piston and cam-driven valves (Three.js) | 1 | Quota refusal | 0.7 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-08 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| gemini-3.1-pro-previewnone (raw API) | Hot air balloon festival over rolling fields (Three.js) | 1 | Quota refusal | 0.4 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-08 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| gemini-3.5-flashnone (raw API) | Hot air balloon festival over rolling fields (Three.js) | 1 | Quota refusal | 0.5 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-08 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |
| Grok 4.7none (raw API) | Hot air balloon festival over rolling fields (Three.js) | 1 | Quota refusal | 0.4 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-08 | WhyThe provider refused the request (HTTP 403: quota, [redacted] or account limit); no reply. | no result yet |
| Kimi K3none (raw API) | Hot air balloon festival over rolling fields (Three.js) | 1 | Quota refusal | 0.5 s | not recorded: the run's token counts were never reported, so there is no estimate | 2026-10-08 | WhyThe provider refused the request (HTTP 429: quota, [redacted] or account limit); no reply. | no result yet |