Pyyan / Compare / Qwen4-Max vs Claude Fable 5.2

Qwen4-Max vs Claude Fable 5.2

2 of 5

Language Models · verified 16 Nov 2026

Qwen4-MaxAlibabacurrent
Claude Fable 5.2Anthropiccurrent
3 slots left
SpecificationQwen4-MaxClaude Fable 5.2
SummaryAlibaba's first Qwen4 model, a closed flagship at $1.50 and $4.50 with a 1M context.Up to 30% faster than 5.1 and a fifth cheaper, with a fast mode at 2.5x.
Context1M1M
Input ($/Mtok)$1.5$8
Output ($/Mtok)$4.5$40
LicenceProprietaryProprietary
Max output131,072 (262,144 thinking plus answer)128K
Parameters2.4T total (reported)Not published
Released16 Nov 20269 Nov 2026
Model IDqwen4-maxclaude-fable-5-2
CategoryLanguage ModelsLanguage Models
OfficialAlibaba ↗Anthropic ↗

Highlighted rows are where these differ.

Qwen4-Max

  • Generally available on 16 November 2026, five days after a Token Plan only preview
  • Reads text, images and video; thinking is on by default, with reasoning effort low, high or xhigh
  • Speaks the OpenAI Chat Completions, OpenAI Responses and Anthropic Messages APIs
  • International list price $1.50 in and $4.50 out per million tokens, against $2 and $6 for Qwen3.8-Max; cache reads $0.075
  • IT Home and Jiemian report 2.4 trillion total parameters; Alibaba says smaller Qwen4 models will be open sourced later

Best for long-context agent and office work at a lower price than Qwen3.8-Max.

Full spec sheet →

Claude Fable 5.2

  • Output up to 30% faster than Fable 5.1 at $8 and $40 per million tokens, against $10 and $50, with an optional fast mode up to 2.5x quicker at $16 and $80
  • 49.1% on FrontierCode 2.0 at top effort in Anthropic's table, against 42.8% for GPT-6 Astra
  • Same model as Mythos 5.2, which is limited to vetted cyber and life sciences organisations. Fable carries the cyber and biology safeguards
  • Anthropic says 1,000 agents formalised the proof of Fermat's Last Theorem in Lean in 30 hours, a result from its own prerelease testing

Best for the hardest long-running technical work.

Full spec sheet →