MODELDRIFT.WATCH

← Day 36 · All days · Day 38 →

GPT, Claude and Gemini had partial days.

No answer came back today, so nothing was compared.

Friday, Oct 16 2026 · Day 37

0 of 150

RESPONSES · EVERY REQUEST FAILED

Missing responses 150

  1. MissingGPT50 of 50 missingNo response: a billing error on our side (HTTP 402, payment required). Missing responses are gaps, never counted as changes.
  2. MissingClaude50 of 50 missingNo response: a billing error on our side (HTTP 402, payment required). Missing responses are gaps, never counted as changes.
  3. MissingGemini50 of 50 missingNo response: a billing error on our side (HTTP 402, payment required). Missing responses are gaps, never counted as changes.
Tokens and costShowHideOutput tokens billed, input tokens, response time and cost, by model
ModelOutput length7 daysOutput tokens billedInput tokensResponse timeCost
GPT··0unchanged·$0.00
Claude··0unchanged·$0.00
Gemini··0unchanged·$0.00
All three0$0.00

Each reading is against the model’s own recent past. The line is output length over the last 30 days, in the band its ordinary day sits in (plus or minus 15%), with a dot on today. A partial day’s cost is what was spent, never scaled up.

All verdictsShowHide150 verdicts, one per question and model, each change marked on its row
ChangePromptModelOutcomeCorrectFormatDetail≈TokensLatencyProviderFlags
D1driftclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.2 s··
D2driftclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
D4driftclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
D5driftclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
D7driftclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
D9driftclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
D13driftclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
D14driftclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
D17driftclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
D20driftclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
D28driftclaude-opus-5error····0.3 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
D34driftclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
R1refusalclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
R5refusalclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
R10refusalclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
R16refusalclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.3 s··
gpt-5.6-solerror····0.1 s··
R19refusalclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
R23refusalclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
R24refusalclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
R30refusalclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
R33refusalclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
R36refusalclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.2 s··
P1recommendationclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
P3recommendationclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
P6recommendationclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.2 s··
P7recommendationclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
P14recommendationclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
P15recommendationclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
P18recommendationclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
P22recommendationclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
P26recommendationclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
P28recommendationclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
P29recommendationclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
P37recommendationclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
F1Aframingclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
F1Bframingclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
F3Aframingclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
F3Bframingclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
F14Aframingclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
F14Bframingclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
F18Aframingclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
F18Bframingclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
F22Aframingclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
F22Bframingclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
H9hedgingclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
H11hedgingclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
H13hedgingclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
H27hedgingclaude-opus-5error····0.2 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
H36hedgingclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··
H37hedgingclaude-opus-5error····0.1 s··
gemini-3.1-pro-previewerror····0.1 s··
gpt-5.6-solerror····0.1 s··

Model names link to the full response. ≈Tokens = visible characters ÷ 4. Latency = wall time of the API call. Detail = first extracted pick, pair stance (arm A|arm B) or scored answer. Definitions.

Run recordShowHideRun, responses, comparison, aliases, versions, triage
Run
20261016T090042Z-e5d9bd
started 2026-10-16 09:00:42 UTC, last response 2026-10-16 09:00:58 UTC
Responses
150 archived · 150 scored · 150 error rows (no data, excluded)
Compared
nothing: no answer came back
Aliases
all 3 aliases resolved to their pinned versions at 2026-10-16 09:00:42 UTC
Versions
scorer v1 · severity rules r2
Triage
0 of 0 ambiguous events triaged

← Day 36 · All days · Day 38 →