{"title":"Cache Economics Beyond the Headline: When Does Fable or Gemini Win on Repeated Context?","description":"The lower cache-read price does not alone determine the lower-cost workflow: cache-write charges, eligibility, hit behavior, expiry, and output tokens can change the result.","dataset_id":"spd:cache-economics-beyond-the-headline-when-does-fable-or-gemini-win-on-repeated-context-b85ac7d5","canonical_url":"https://superpowerdaily.com/research/cache-economics-beyond-the-headline-when-does-fable-or-gemini-win-on-repeated-context-b85ac7d5","version_url":"https://superpowerdaily.com/research/cache-economics-beyond-the-headline-when-does-fable-or-gemini-win-on-repeated-context-b85ac7d5","version":"v2","snapshot_hash":"5f105f388641b55f833dcb3f700ed37c246738a2110647e6c8db497445fc75c2","date_created":"2026-09-03T17:32:39.549Z","date_modified":"2026-09-04T17:36:13.283Z","license":{"name":"Superpower Daily data reuse terms","url":"https://superpowerdaily.com/terms"},"license_url":"https://superpowerdaily.com/terms","temporal_coverage":"2026-09-03","coverage_note":"This synthesis uses primary provider documentation and derived calculations from the supplied rates. Artifact mode is evidence_matrix, so no chart is included.","measurement_technique":["Evidence matrix plan: reconcile provider price documentation, cache eligibility and expiry rules, API usage fields, and organization-level billing records where available.","Compare only repeated-prefix input costs under equal billed prefix-token counts; exclude output and changing suffix tokens from crossover calculations.","Use reported provider token fields for invoice reconciliation because the same content can tokenize differently between providers.","Model Gemini implicit-cache behavior as a hit fraction rather than a guaranteed TTL-based outcome.","Treat Fable cache writes as expiry clusters: a new cluster is required after expiry or invalidation."],"methodology":["Evidence matrix plan: reconcile provider price documentation, cache eligibility and expiry rules, API usage fields, and organization-level billing records where available.","Compare only repeated-prefix input costs under equal billed prefix-token counts; exclude output and changing suffix tokens from crossover calculations.","Use reported provider token fields for invoice reconciliation because the same content can tokenize differently between providers.","Model Gemini implicit-cache behavior as a hit fraction rather than a guaranteed TTL-based outcome.","Treat Fable cache writes as expiry clusters: a new cluster is required after expiry or invalidation."],"metrics":[{"label":"Verified observations","value":"24","detail":"23 measured fields"},{"label":"Supported claims","value":"9","detail":"9 material findings"},{"label":"Cited sources","value":"11","detail":"11 primary or authoritative"},{"label":"Research score","value":"87","detail":"Automated topic and evidence score"}],"columns":[{"key":"entity","label":"Entity"},{"key":"metric","label":"Metric"},{"key":"value","label":"Value"},{"key":"unit","label":"Unit"},{"key":"observed","label":"Observed"},{"key":"source","label":"Source"},{"key":"transform","label":"Transform"}],"data":[{"unit":"USD","value":"$1.875","entity":"Claude Fable 5.1","metric":"100,000-token prefix cost, five-minute write plus 25 hits","source":"https://platform.claude.com/docs/en/models/fable-5-1/overview","observed":"2026-09-03","transform":"0.1*(12.5+0.25*25)."},{"unit":"USD","value":"$0.2625","entity":"Gemini 3.8 Flash","metric":"100,000-token prefix cost, initial request plus 25 implicit hits","source":"https://ai.google.dev/gemini-api/docs/pricing","observed":"2026-09-03","transform":"0.1*(0.75+0.075*25)."},{"unit":"USD","value":"$1.95","entity":"Gemini 3.8 Flash","metric":"100,000-token prefix cost, initial request plus 25 implicit misses","source":"https://ai.google.dev/gemini-api/docs/pricing","observed":"2026-09-03","transform":"0.1*(0.75+0.75*25)."},{"unit":"USD/MTok","value":"$0.075 per million tokens","entity":"Gemini 3.8 Flash","metric":"cached-context price through 2026-12-31","source":"https://ai.google.dev/gemini-api/docs/pricing","observed":"2026-09-03","transform":null},{"unit":"USD/MTok","value":"$0.25 per million tokens","entity":"Claude Fable 5.1","metric":"cache-read price","source":"https://platform.claude.com/docs/en/models/fable-5-1/overview","observed":"2026-09-03","transform":null},{"unit":"USD/MTok","value":"$12.50 per million tokens","entity":"Claude Fable 5.1","metric":"five-minute cache-write price","source":"https://platform.claude.com/docs/en/models/fable-5-1/overview","observed":"2026-09-03","transform":null},{"unit":"percent of reuse tokens","value":"74.074%","entity":"Claude Fable 5.1 versus Gemini 3.8 Flash","metric":"Gemini hit fraction at which Fable loses for every reuse count","source":"https://ai.google.dev/gemini-api/docs/pricing","observed":"2026-09-03","transform":"Solved 0.75*(1-h)+0.075*h=0.25; h=0.5/0.675."},{"unit":"tokens","value":"512 tokens","entity":"Claude Fable 5.1","metric":"minimum cacheable prefix","source":"https://platform.claude.com/docs/en/build-with-claude/prompt-caching","observed":"2026-09-03","transform":null},{"unit":"tokens","value":"4,096 tokens","entity":"Gemini 3.8 Flash","metric":"minimum implicit-cache prefix","source":"https://ai.google.dev/gemini-api/docs/caching","observed":"2026-09-03","transform":null},{"unit":"reuse requests","value":"24 reuses","entity":"Claude Fable 5.1 versus Gemini 3.8 Flash","metric":"minimum reuses, five-minute Fable cache, 0% Gemini hits","source":"https://platform.claude.com/docs/en/models/fable-5-1/overview","observed":"2026-09-03","transform":"floor(11.75/(0.5-0.675*0))+1."},{"unit":"reuse requests","value":"36 reuses","entity":"Claude Fable 5.1 versus Gemini 3.8 Flash","metric":"minimum reuses, five-minute Fable cache, 25% Gemini hits","source":"https://platform.claude.com/docs/en/models/fable-5-1/overview","observed":"2026-09-03","transform":"floor(11.75/(0.5-0.675*0.25))+1."},{"unit":"reuse requests","value":"73 reuses","entity":"Claude Fable 5.1 versus Gemini 3.8 Flash","metric":"minimum reuses, five-minute Fable cache, 50% Gemini hits","source":"https://platform.claude.com/docs/en/models/fable-5-1/overview","observed":"2026-09-03","transform":"floor(11.75/(0.5-0.675*0.5))+1."},{"unit":"reuse requests","value":"124 reuses","entity":"Claude Fable 5.1 versus Gemini 3.8 Flash","metric":"minimum reuses, five-minute Fable cache, 60% Gemini hits","source":"https://platform.claude.com/docs/en/models/fable-5-1/overview","observed":"2026-09-03","transform":"floor(11.75/(0.5-0.675*0.6))+1."},{"unit":"reuse requests","value":"428 reuses","entity":"Claude Fable 5.1 versus Gemini 3.8 Flash","metric":"minimum reuses, five-minute Fable cache, 70% Gemini hits","source":"https://platform.claude.com/docs/en/models/fable-5-1/overview","observed":"2026-09-03","transform":"floor(11.75/(0.5-0.675*0.7))+1."},{"unit":"reuse requests","value":"39 reuses","entity":"Claude Fable 5.1 versus Gemini 3.8 Flash","metric":"minimum reuses, one-hour Fable cache, 0% Gemini hits","source":"https://platform.claude.com/docs/en/models/fable-5-1/overview","observed":"2026-09-03","transform":"floor(19.25/(0.5-0.675*0))+1."},{"unit":"reuse requests","value":"59 reuses","entity":"Claude Fable 5.1 versus Gemini 3.8 Flash","metric":"minimum reuses, one-hour Fable cache, 25% Gemini hits","source":"https://platform.claude.com/docs/en/models/fable-5-1/overview","observed":"2026-09-03","transform":"floor(19.25/(0.5-0.675*0.25))+1."},{"unit":"reuse requests","value":"119 reuses","entity":"Claude Fable 5.1 versus Gemini 3.8 Flash","metric":"minimum reuses, one-hour Fable cache, 50% Gemini hits","source":"https://platform.claude.com/docs/en/models/fable-5-1/overview","observed":"2026-09-03","transform":"floor(19.25/(0.5-0.675*0.5))+1."},{"unit":"reuse requests","value":"203 reuses","entity":"Claude Fable 5.1 versus Gemini 3.8 Flash","metric":"minimum reuses, one-hour Fable cache, 60% Gemini hits","source":"https://platform.claude.com/docs/en/models/fable-5-1/overview","observed":"2026-09-03","transform":"floor(19.25/(0.5-0.675*0.6))+1."},{"unit":"reuse requests","value":"701 reuses","entity":"Claude Fable 5.1 versus Gemini 3.8 Flash","metric":"minimum reuses, one-hour Fable cache, 70% Gemini hits","source":"https://platform.claude.com/docs/en/models/fable-5-1/overview","observed":"2026-09-03","transform":"floor(19.25/(0.5-0.675*0.7))+1; strict inequality requires 701 rather than 700."},{"unit":"USD/MTok","value":"$20 per million tokens","entity":"Claude Fable 5.1","metric":"one-hour cache-write price","source":"https://platform.claude.com/docs/en/models/fable-5-1/overview","observed":"2026-09-03","transform":null},{"unit":"USD when C is millions of tokens","value":"C*(0.25*N+19.75*K)","entity":"Claude Fable 5.1","metric":"prefix-cost formula with expiry clusters","source":"https://platform.claude.com/docs/en/build-with-claude/prompt-caching","observed":"2026-09-03","transform":"Expanded C*(20*K+0.25*(N-K)); N is total requests and K is cache-write clusters."},{"unit":"USD when C is millions of tokens","value":"C*(0.25*N+12.25*K)","entity":"Claude Fable 5.1","metric":"prefix-cost formula with expiry clusters","source":"https://platform.claude.com/docs/en/build-with-claude/prompt-caching","observed":"2026-09-03","transform":"Expanded C*(12.5*K+0.25*(N-K)); N is total requests and K is cache-write clusters."},{"unit":"USD/MTok","value":"$0.75 per million tokens","entity":"Gemini 3.8 Flash","metric":"standard input price through 2026-12-31","source":"https://ai.google.dev/gemini-api/docs/pricing","observed":"2026-09-03","transform":null},{"unit":"USD/MTok","value":"$10 per million tokens","entity":"Claude Fable 5.1","metric":"uncached input price","source":"https://platform.claude.com/docs/en/models/fable-5-1/overview","observed":"2026-09-03","transform":null}],"sources":[{"url":"https://ai.google.dev/gemini-api/docs/caching","name":"Google","title":"Context caching  |  Gemini API  |  Google AI for Developers","records":1},{"url":"https://ai.google.dev/gemini-api/docs/generate-content/caching","name":"Google","title":"Context caching  |  Gemini Generate Content API (Legacy)  |  Google AI for Developers","records":0},{"url":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/context-cache/context-cache-overview","name":"Google Cloud","title":"Context caching overview  |  Gemini Enterprise Agent Platform  |  Google Cloud Documentation","records":0},{"url":"https://ai.google.dev/api/tokens","name":"Google","title":"Counting tokens  |  Gemini API  |  Google AI for Developers","records":0},{"url":"https://platform.claude.com/docs/en/api/messages/create","name":"Anthropic","title":"Create a Message - Claude API Reference","records":0},{"url":"https://ai.google.dev/gemini-api/docs/pricing","name":"Google","title":"Gemini Developer API pricing  |  Gemini API  |  Google AI for Developers","records":5},{"url":"https://platform.claude.com/docs/en/models/fable-5-1/overview","name":"Anthropic","title":"platform.claude.com","records":15},{"url":"https://platform.claude.com/docs/en/about-claude/pricing","name":"Anthropic","title":"platform.claude.com","records":0},{"url":"https://platform.claude.com/docs/en/build-with-claude/prompt-caching","name":"Anthropic","title":"platform.claude.com","records":3},{"url":"https://platform.claude.com/docs/en/manage-claude/usage-cost-api","name":"Anthropic","title":"Usage and Cost API","records":0},{"url":"https://ai.google.dev/gemini-api/docs/zdr","name":"Google","title":"Zero data retention in the Gemini Developer API  |  Gemini API  |  Google AI for Developers","records":0}],"provenance":{"publisher":"Superpower Daily","source_count":11,"source_urls":["https://ai.google.dev/gemini-api/docs/caching","https://ai.google.dev/gemini-api/docs/generate-content/caching","https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/context-cache/context-cache-overview","https://ai.google.dev/api/tokens","https://platform.claude.com/docs/en/api/messages/create","https://ai.google.dev/gemini-api/docs/pricing","https://platform.claude.com/docs/en/models/fable-5-1/overview","https://platform.claude.com/docs/en/about-claude/pricing","https://platform.claude.com/docs/en/build-with-claude/prompt-caching","https://platform.claude.com/docs/en/manage-claude/usage-cost-api","https://ai.google.dev/gemini-api/docs/zdr"],"methodology":["Evidence matrix plan: reconcile provider price documentation, cache eligibility and expiry rules, API usage fields, and organization-level billing records where available.","Compare only repeated-prefix input costs under equal billed prefix-token counts; exclude output and changing suffix tokens from crossover calculations.","Use reported provider token fields for invoice reconciliation because the same content can tokenize differently between providers.","Model Gemini implicit-cache behavior as a hit fraction rather than a guaranteed TTL-based outcome.","Treat Fable cache writes as expiry clusters: a new cluster is required after expiry or invalidation."],"snapshot_hash":"5f105f388641b55f833dcb3f700ed37c246738a2110647e6c8db497445fc75c2"},"distributions":{"csv":"https://superpowerdaily.com/api/research/cache-economics-beyond-the-headline-when-does-fable-or-gemini-win-on-repeated-context-b85ac7d5?format=csv","json":"https://superpowerdaily.com/api/research/cache-economics-beyond-the-headline-when-does-fable-or-gemini-win-on-repeated-context-b85ac7d5?format=json"}}