claude-fable-5 deep verification model-consistency report
Report ID MZZR202607090325405F2410Generated at 2026/07/09 11:25
Matched with issues
claude-fable-5 deep verification result: Matched with issues (79/100). Main deductions: API service error; Model-specific boundary response was unclear.
79Consistency score
主要扣分项：接口服务异常；模型专属边界表达不清。
EvaluatorMozhenzhen
Tested modelClaude Fable 5
ProvidersInfistar
Samples30 probes
Duration3 min 11 sec
Verification versionV11
Identity verdictVerified
Verification basisMozhenzhen versioned baseline
Token usage
Measured API usage is shown first; official baselines or test estimates are used only when measured values are unavailable.
Measured token usage239.0KFrom the response usage fields; this is the report's primary usage measure.Test-suite estimate74.9KPrecisely estimated from this test suite when no official baseline is available.Variance assessment219% aboveMeasured token usage is above the estimate; review billing or provider logs.Usage conclusionAbove estimateThis reflects usage observability, not the final billed amount.
Token cache usage test
Observes token usage, cache fields, and reuse across rounds; 0% means no cached-token field was observed.
Cache share by testRepresents the API-reported cache share, not the final billing discount.
Long-context test
Shows stable responses at different context lengths and the model center reference window.
1 samples
边界在 32K
32K 需要关注，原因：中继链路不可用。
Reference windowReference window 1MModel-center context window; available lengths may vary by provider.Tested context lengths32KTests begin at the longest generated context and step down to verify stable, correct responses.
Current guidance
32K 档暂不作为当前线路的稳定上下文参考。
A newer report is availableThis report has been superseded by a later retest. View the latest report。
Model authenticity score
Review API availability, model identity, response completeness, and other checks separately.
Scoring breakdown
Matched with issues79/100
主要扣分项：接口服务异常；模型专属边界表达不清。
Callability and model visibility10 / 10
No deductions
Protocol and response structure8 / 12
接口服务异常（标准题：usage 返回可观测性） -4
Identity and model match12 / 18
模型专属边界表达不清（深度模型题：模型族专属边界） -6
Core capabilities and model-specific tests22 / 25
模型专属边界表达不清（深度模型题：模型族专属边界） -3
Usage and billing consistency10 / 10
No deductions
Reliability and availability17 / 25
接口服务异常（标准题：usage 返回可观测性） -8
Test details
Expand each check to view the redacted evidence notes.
18 checks
Model list visibility
claude-fable-5 is visible in the API model list.
Passed
Test evidence
Check: Model list visibility
Status: Passed
Samples: 8
API path: /v1/models
Visible models: 8
Basic call probe
The basic API call succeeded and returned a recognizable response.
claude-fable-5 deep verification result: Matched with issues (79/100). Main deductions: API service error; Model-specific boundary response was unclear.
EvaluatorMozhenzhen
Tested modelClaude Fable 5
ProvidersInfistar
Samples30
Duration3 min
Test versionV11
Identity verdictVerified
Verification basisMozhenzhen versioned baseline
Token usage
Measured API usage is shown first; official baselines or test estimates are used only when measured values are unavailable.
Measured token usage239.0KFrom the response usage fields; this is the report's primary usage measure.Test-suite estimate74.9KPrecisely estimated from this test suite when no official baseline is available.Variance assessment219% aboveMeasured token usage is above the estimate; review billing or provider logs.Usage conclusionAbove estimateThis reflects usage observability, not the final billed amount.
Token cache usage test
Observes token usage, cache fields, and reuse across rounds; 0% means no cached-token field was observed.
Cache share by testRepresents the API-reported cache share, not the final billing discount.
Long-context test
Shows stable responses at different context lengths and the model center reference window.
1 samples
边界在 32K
32K 需要关注，原因：中继链路不可用。
Reference windowReference window 1MModel-center context window; available lengths may vary by provider.Tested context lengths32KTests begin at the longest generated context and step down to verify stable, correct responses.
Current guidance
32K 档暂不作为当前线路的稳定上下文参考。
A newer report is availableThis report has been superseded by a later retest. View the latest report。
Model authenticity score
Review API availability, model identity, response completeness, and other checks separately.
Scoring breakdown
Matched with issues79/100
主要扣分项：接口服务异常；模型专属边界表达不清。
Callability and model visibility10 / 10
No deductions
Protocol and response structure8 / 12
接口服务异常（标准题：usage 返回可观测性） -4
Identity and model match12 / 18
模型专属边界表达不清（深度模型题：模型族专属边界） -6
Core capabilities and model-specific tests22 / 25
模型专属边界表达不清（深度模型题：模型族专属边界） -3
Usage and billing consistency10 / 10
No deductions
Reliability and availability17 / 25
接口服务异常（标准题：usage 返回可观测性） -8
Test details
Expand each check to view the redacted evidence notes.
18 checks
Model list visibility
claude-fable-5 is visible in the API model list.
Passed
Test evidence
Check: Model list visibility
Status: Passed
Samples: 8
API path: /v1/models
Visible models: 8
Basic call probe
The basic API call succeeded and returned a recognizable response.
