Claude Fable 5
Anthropic · Lançado 06/2026
Anthropic's Mythos-class frontier model, the strongest publicly available Claude at its June 2026 launch, now listed as legacy behind Claude Fable 5.1. The verdict stays conditional on two constraints: API pricing of $10/$50 per 1M tokens, and a safety classifier that over-triggered on routine coding tasks at launch, routing them to an Opus 4.8 fallback. Subscribers reach Fable through Claude Max at 50% of weekly limits, or through usage credits on Pro. Anthropic's current lineup lists Fable 5.1 and Opus 5.5 for new work.
É adequada para ti?
Bom para
- Codificação autónoma de longo horizonte — comprime meses de engenharia em dias (Stripe: migração de 50M de linhas em 1 dia contra um esforço de equipa de 2 meses)
- Raciocínio analítico complexo — Hebbia Finance Benchmark #1, geração de hipóteses ao nível de cientista de investigação sénior
- Tarefas intensivas em visão — interpretação de gráficos/tabelas de última geração, reconstrói aplicações web a partir de capturas de ecrã, completa Pokémon FireRed apenas com visão
- Fluxos de trabalho agênticos com memória persistente — executa tarefas multi-turno de forma autónoma com memória baseada em ficheiros, autocorrige-se a partir de notas
Não recomendado para
- Consultas de cibersegurança e biologia — os classificadores de segurança recorrem ao Opus 4.8; trabalho legítimo de segurança/ciências da vida esbarra na barreira de salvaguarda
- Produção de alto volume sensível ao custo — a $50/1M de saída, os custos de processamento em lote agravam-se rapidamente face ao GPT-5.5 a $30 ou ao Opus 4.8 a $25
- Ambientes regulados com restrições de residência de dados — a retenção obrigatória de dados de 30 dias para modelos da classe Mythos é um fardo de conformidade
Desempenho por tarefa
Geração de código
Líder da classe: 80,3% SWE-bench Pro, #1 FrontierCode, validado pela Stripe em escala de produção
Raciocínio de contexto longo
Mantém o foco ao longo de milhões de tokens; a memória persistente melhora o desempenho em jogos 3x face ao Opus 4.8
Trabalho de conhecimento
Melhores pontuações em benchmarks financeiros/analíticos; forte em revisão jurídica, análise de trading, raciocínio de causa-raiz
Visão
Novo estado da arte — extrai números precisos de figuras científicas, conclusão de jogos apenas com visão
Investigação científica
Produz novas hipóteses de biologia molecular (80% de preferência dos cientistas sobre a classe Opus); aceleração de 10x no design de fármacos
Cibersegurança (ofensiva)
Mythos 5 atinge 78% no ExploitBench, mas o Fable 5 está bloqueado pelos classificadores — recorre ao Opus 4.8
Preços
Entrada
$10 / 1M tokens
Saída
$50 / 1M tokens
Contexto
1M ctx, 128K out; legacy; Max plans at 50% limits
Testes de desempenho
Histórico de veredictos
Fontes
- Anthropic Claude pricing page10/2026
- Weights & Biases: Fable 5 Benchmark Scores06/2026
- Anthropic pricing page10/2026
- Anthropic Claude Fable 5 and Mythos 5 announcement10/2026
- OpenRouter: Claude Fable 5 API Pricing & Benchmarks06/2026
- Anthropic: System card06/2026
- Anthropic: Claude Fable product page06/2026
- LLM Stats: Claude Fable 5 Review, Benchmarks and Pricing06/2026
- Let's Data Science: Anthropic Restores Fable 5 After US Ban06/2026
- Anthropic — Statement on US government directive06/2026
- Claude API docs: Models overview (Sonnet 4.6 and Sonnet 5 listed as legacy, still available; Sonnet 5.5 current)10/2026
Registo de verificação
- Preços· Sem alterações
Agente automatizado
- Preços· Sem alterações
Agente automatizado
- Preços· Sem alterações
Agente automatizado
- Conteúdo· Atualizado
Agente automatizado
Summary said credit-only with no subscription path; claude.com/pricing lists Fable at 50% of weekly limits on Max and via usage credits on Pro. Free has Sonnet and Haiku only. Team row conflicting and not written. Rate $10/$50 unchanged; Fable 5 is legacy behind Fable 5.1. Summary corrected via verdict_set (conditional held); sources and pricing_context refreshed.
- Preços· Sem alterações
Agente automatizado
- Preços· Sem alterações
Agente automatizado
Claude Fable 5 $10 in / $50 out, cache hits $1, batch $5/$25, no limited-availability tag. Matches stored scalars. No update. verdict_summary 'no subscription path' still contradicts pricing_context 'Max plans at 50% limits'. The fix rides held draft claude-fable-5-max-plan-access-correction. Not re-staged. claude.com/pricing (plan access) was not re-read this run.
- Perfil· Sem alterações
Agente automatizado
Rate card: Claude Fable 5 $10/$50, cache hits $1, batch $5/$25, no limited-availability tag; matches stored. Individual plan row 'Fable': Free No, Pro Usage credits, Max 5x / Max 20x '50% of weekly limits*'. Matches pricing_context and audiences. TEAM ACCESS UNRESOLVED, two reads disagree: a first summarised fetch of claude.com/pricing reported Team Standard/Premium 'Fable 50% of weekly limits* on premium seats'; a second verbatim-constrained fetch minutes later returned 'NO FABLE ROW IN TEAM SECTION' (seat prices Standard $20/$25, Premium $100/$125). Consistent with 09-09/09-14 reads (no Team row). Nothing written about Team. Summary 'no subscription path' is still false; its fix rides held draft claude-fable-5-max-plan-access-correction (verdict_history latest still 2026-07-15). Not re-staged, no duplicate draft. No update.
- Perfil· Sem alterações
Agente automatizado
Access re-read: individual plan table row labelled exactly 'Fable' (no version number): Free 'No', Pro 'Usage credits', Max 5x and Max 20x '50% of weekly limits*' (footnote 'Usage limits apply.'); Fable absent from the Team and Enterprise tables as the summarizer read them. Rate card: Claude Fable 5 $10/$50, cache hits $1, batch $5/$25, no limited-availability tag. Stored scalars and pricing_context 'Max plans at 50% limits' match. No update. The summary's 'no subscription path' is still false against this page; its fix rides held draft claude-fable-5-max-plan-access-correction (payload not landed, verdict_history latest 2026-07-15). Not re-staged, no duplicate draft. Still unresolved: which Fable version the unversioned Max row grants (also the open claude-fable-5-1 question). Not guessed.
- Perfil· Atualizado
Agente automatizado
Re-verification of the access finding first logged 2026-09-07, confirmed independently against the live page today. Anthropic's plan table lists Claude Fable as: Free not available, Pro usage credits, Max 5x and Max 20x 50% of weekly limits, Team not available, Enterprise (self-serve and sales-assisted) available. A subscription path therefore EXISTS, and the stored verdict_summary claim 'credit-only pricing at $10/$50 per 1M tokens with no subscription path' is false. API rates re-verified UNCHANGED on the platform rate card: $10 in / $50 out per 1M, cache reads $1 per 1M; Fable 5 carries no limited-availability tag, unlike Mythos 5 and Mythos 5.1 which link to anthropic.com/glasswing. ONE field changed this run: pricing_url moved from https://www.anthropic.com/pricing (not fetched, unverified) to the platform rate card actually read. The structured children were already corrected on 2026-09-07. THE SUMMARY WAS NOT, AND THIS IS THE FINDING: the 2026-09-07 pass staged the rewrite on a draft named 'claude-fable-5-max-plan-access-correction'; that draft EXISTS (post_create returned 409 duplicate slug this run) but was never published, so the verdict never moved and the entity has continued to state the false half in the field readers quote while its own pricing_context and audience rows state the true half. No duplicate draft created. The entity_verdict payload for the EXISTING draft is staged in this run's return. Rating held at conditional on the surviving safety-classifier condition. NOT ESTABLISHED, deliberately left open: the plan table says 'Fable' with no version number while the API card lists Fable 5 and Fable 5.1 separately, so which version a Max subscriber reaches is unresolved and was not guessed.
- Perfil· Atualizado
Agente automatizado
Resolves the carried access trigger. API rates re-verified UNCHANGED at $10 in / $50 out per 1M with cache reads at $1 per 1M (platform.claude.com pricing table). ACCESS CHANGED against stored: the verdict summary asserts 'credit-only pricing ... with no subscription path', but Anthropic's live plan table lists Claude Fable as Free: not available, Pro: usage credits, Max 5x and Max 20x: 50% of weekly limits. A Max subscription path exists; the credit-only description holds for Pro only. This confirms our own W29 published record (Fable 5 added to Max/Team from Jul 20) against a live primary, closing a stale-self-authored-prose drift open since 2026-07-15. Added a Max-subscriber for_whom audience and a Free-tier not_for_whom, refreshed pricing_context to note Max at 50% limits. Fable 5 carries NO 'limited availability' tag on the pricing table (unlike Mythos 5 and Mythos 5.1). Rating held at conditional on the surviving safety-classifier condition; verdict summary rewrite staged for the announcing draft 'claude-fable-5-max-plan-access-correction'.
- Preços· Sem alterações
Agente automatizado
- Preços· Sem alterações
Agente automatizado
- Preços· Sem alterações
Agente automatizado
- Preços· Sem alterações
Agente automatizado
- Estado· Atualizado
Agente automatizado
Summary re-dated to Jul 7 2026: added Jul 1 return, safety-router over-blocking (silent Opus 4.8 fallback), and the Jul 7 credit cliff (credit-only $10/$50 from Jul 8, no auto fallback, no return timeline). Verdict stays conditional; condition restated as credit-only access + over-triggering safety router.
- Preços· Sem alterações
Agente automatizado
- Preços· Sem alterações
Agente automatizado
Pricing unchanged: $10/$50 per 1M tokens. Anthropic pricing page STILL shows 'Claude Fable 5 is unavailable.' CryptoBriefing reports ban lift but independent tracker isfable5back.com confirms API still suspended. Ban lift NOT reflected on primary source (Anthropic pricing page).
- Perfil· Sem alterações
Importado no lançamento
- Preços· Sem alterações
Importado no lançamento