Choose Claude Sonnet 5 When
- $2 input and $10 output per million tokens, versus $10 and $50
- A 71.5 Coding Index, compared with 62.0 for Claude Fable 5
- The same 1M context, 128K output, and text, image, and PDF inputs
Claude Sonnet 5 vs Claude Fable 5
Claude Sonnet 5 from Anthropic and Claude Fable 5 from Anthropic are shown side by side so you can compare pricing, model IDs, context and output limits, modalities, tool use, release details, and benchmark results where data is available.
Overall Recommendation
Claude Sonnet 5 is the stronger production value default: it costs 80% less per token and has a 71.5 Coding Index versus 62.0 for Fable 5. Choose Fable 5 when its higher Intelligence and HLE scores or Mythos-level long-running project focus matter more.
Recommendation based on Claude Sonnet 5 (Non-reasoning, High Effort) vs Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback).
Comparison Highlights
Claude Sonnet 5Anthropic
Claude Fable 5Anthropic
Best For
Cost-efficient codingscalable agentsbalanced production work
Ambitious long-running projectsautonomous knowledge worksubagent collaboration
How To Read ItUse-case synthesis from model capabilities and positioning.
Workload Fit
CodingScaleValue
Knowledge workLong projectsSubagents
How To Read ItThe workloads each model is most directly shaped for.
Input
$2/MEdge
$10/M
How To Read ItLower is better when prompts or retrieval payloads are large.
Output
$10/MEdge
$50/M
How To Read ItLower is better for reasoning-heavy or long-generation traffic.
Context length
1M
1M
How To Read ItHigher is better for repositories, retrieval packs, and transcripts.
Input modalities
Text, Image, PDF
Text, Image, PDF
How To Read ItBroader input support reduces the need for separate vision or video models.
Model Selection Guide
Choose Claude Fable 5 When
- A 59.9 Intelligence Index, versus 53.4 for Claude Sonnet 5
- A 53.3 HLE score, compared with 39.6 for Sonnet 5
- Mythos-level positioning for memory-oriented, long-running project work
Basic Information
Claude Sonnet 5Anthropic
Claude Fable 5Anthropic
Name
Claude Sonnet 5Claude Sonnet 5
Model id
Claude Sonnet 5claude-sonnet-5
Intro
Claude Sonnet 5Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work. It supports adaptive thinking with selectable reasoning effort levels (low, medium, high, max, and x-high), a 1M-token context window, and text, image, and file inputs. Sonnet 5 uses an updated tokenizer and includes real-time cyber safeguards that block certain high-risk dual-use activities.
Author
Claude Sonnet 5Anthropic
Released date
Claude Sonnet 52026-06-30
Context length
Claude Sonnet 51M
Max output tokens
Claude Sonnet 5128K
Name
Claude Fable 5Claude Fable 5
Model id
Claude Fable 5claude-fable-5
Intro
Claude Fable 5Claude Fable 5 is presented as a Mythos-level Claude model for ambitious, long-running projects. Source pages emphasize autonomous knowledge work, software engineering, vision, memory, and the ability to work for extended periods with sub-agents. Its description should feel more like a project-level collaborator than a short-turn assistant.
Author
Claude Fable 5Anthropic
Released date
Claude Fable 52026-06-09
Context length
Claude Fable 51M
Max output tokens
Claude Fable 5128K
Name
Claude Sonnet 5Claude Sonnet 5
Name
Claude Fable 5Claude Fable 5
Model id
Claude Sonnet 5claude-sonnet-5
Model id
Claude Fable 5claude-fable-5
Intro
Claude Sonnet 5Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work. It supports adaptive thinking with selectable reasoning effort levels (low, medium, high, max, and x-high), a 1M-token context window, and text, image, and file inputs. Sonnet 5 uses an updated tokenizer and includes real-time cyber safeguards that block certain high-risk dual-use activities.
Intro
Claude Fable 5Claude Fable 5 is presented as a Mythos-level Claude model for ambitious, long-running projects. Source pages emphasize autonomous knowledge work, software engineering, vision, memory, and the ability to work for extended periods with sub-agents. Its description should feel more like a project-level collaborator than a short-turn assistant.
Author
Claude Sonnet 5Anthropic
Author
Claude Fable 5Anthropic
Released date
Claude Sonnet 52026-06-30
Released date
Claude Fable 52026-06-09
Context length
Claude Sonnet 51M
Context length
Claude Fable 51M
Max output tokens
Claude Sonnet 5128K
Max output tokens
Claude Fable 5128K
Pricing
Claude Sonnet 5Anthropic
Claude Fable 5Anthropic
Input
Claude Sonnet 5$2/M
Output
Claude Sonnet 5$10/M
Cached input
Claude Sonnet 5$0.2/M
Input
Claude Fable 5$10/M
Output
Claude Fable 5$50/M
Cached input
Claude Fable 5$1/M
Input
Claude Sonnet 5$2/M
Input
Claude Fable 5$10/M
Output
Claude Sonnet 5$10/M
Output
Claude Fable 5$50/M
Cached input
Claude Sonnet 5$0.2/M
Cached input
Claude Fable 5$1/M
Capabilities
Claude Sonnet 5Anthropic
Claude Fable 5Anthropic
Reasoning
Claude Sonnet 5
Knowledge
Claude Sonnet 52026-01-31
Attachment
Claude Sonnet 5
Input modalities
Claude Sonnet 5Text, Image, PDF
Output modalities
Claude Sonnet 5Text
Temperature
Claude Sonnet 5
Tool use
Claude Sonnet 5
Reasoning
Claude Fable 5
Knowledge
Claude Fable 52026-01-31
Attachment
Claude Fable 5
Input modalities
Claude Fable 5Text, Image, PDF
Output modalities
Claude Fable 5Text
Temperature
Claude Fable 5
Tool use
Claude Fable 5
Reasoning
Claude Sonnet 5
Reasoning
Claude Fable 5
Knowledge
Claude Sonnet 52026-01-31
Knowledge
Claude Fable 52026-01-31
Attachment
Claude Sonnet 5
Attachment
Claude Fable 5
Input modalities
Claude Sonnet 5Text, Image, PDF
Input modalities
Claude Fable 5Text, Image, PDF
Output modalities
Claude Sonnet 5Text
Output modalities
Claude Fable 5Text
Temperature
Claude Sonnet 5
Temperature
Claude Fable 5
Tool use
Claude Sonnet 5
Tool use
Claude Fable 5
Benchmark
Claude Sonnet 5Anthropic
Claude Fable 5Anthropic
Intelligence
41.7
Coding
wins66.4
Intelligence
wins59.9
Coding
62
Knowledge & Reasoning
GPQA
80%
HLE
17.8%
GPQA
wins92.6%
HLE
wins53.3%
Coding
SciCode
48.6%
Terminal-Bench Hard
n/a
SciCode
wins60.2%
Terminal-Bench Hard
wins62.9%
Instruction Following & Agent Tasks
IFBench
n/a
AA-LCR
58.7%
Tau2
n/a
IFBench
wins63.5%
AA-LCR
wins70%
Tau2
wins98.5%
FAQ
Which model is better, Claude Sonnet 5 or Claude Fable 5?
+
Claude Sonnet 5 and Claude Fable 5 should be compared by workload. This page places price, context, output limits, capabilities, and benchmark data side by side. It is also relevant to: claude sonnet 5 vs claude fable 5, sonnet 5 vs fable 5, claude sonnet vs fable.
Is Claude Fable 5 cheaper than Claude Sonnet 5?
+
Use the pricing rows above to compare input, output, and cached input prices for Claude Sonnet 5 and Claude Fable 5.
Which model supports a longer context length?
+
Claude Sonnet 5 lists 1M context length, while Claude Fable 5 lists 1M.
Can I access both models through TokenHub?
+
If both models are available in the TokenHub catalog, you can route requests to Claude Sonnet 5 and Claude Fable 5 through the TokenHub API.