Plain-English Explainer
Anthropic Model Names Explained: What Fable, Opus, Sonnet, and Haiku Actually Mean
Every name has a job. Here is what they are.
Anthropic names its models with two layers: a class name and a tier name. The class (like Mythos) signals a generation leap, while the tier (Opus, Sonnet, Haiku) tells you where a model sits on the capability-versus-speed spectrum. This page decodes both layers, covers the current lineup as of mid-2026, and explains what you are actually running when you open the Claude app.
Anthropic organizes Claude models by class and tier. The Mythos class is the current generation, and Fable 5 (released June 9, 2026) is its first publicly available model. Within any class, Opus is the flagship tier, Sonnet is the balanced mid-tier, and Haiku is the fast lightweight tier. Current versions are Opus 4.8, Sonnet 4.6, and Haiku 4.5. The number after the tier name is a version counter, not a separate family.
The Tier System
Three Tiers, One Spectrum
Every Claude model sits in one of three tiers. The tier tells you the capability-versus-speed trade-off before you read anything else.
Opus: the flagship tier
Highest capability, deepest reasoning. Opus 4.8 is the current model. Use it for complex analysis, architecture decisions, and tasks where accuracy matters more than speed.
Sonnet: the balanced tier
Sonnet 4.6 hits the middle of the spectrum. It handles most real-world coding, writing, and research tasks well without the latency of a flagship model.
Haiku: the fast tier
Haiku 4.5 is built for speed and volume. It suits high-frequency tasks like classification, summarization, and lightweight tool calls where low latency is the priority.
Mythos: the current class
Mythos is the generation name. Fable 5 is the first Mythos model released to the public. Earlier Claude models used a different internal class designation.
Fable 5: the Mythos debut
Fable 5 launched June 9, 2026. It was the first model from the Mythos class to leave Anthropic's internal research and become publicly available.
How the Names Work Together
Reading a Model Name End to End
A full Claude model name has two parts: the tier name and a version number. 'Sonnet 4.6' means the Sonnet tier, version 4.6. The version counter goes up when Anthropic ships an update to that tier. It is not a separate product family.
The class name sits above the tier. Mythos is the current class, and Fable 5 is its first publicly available model. Class names signal a generational leap, so when a new class name appears in an announcement, that is Anthropic telling you this is not just a version bump.
When Anthropic announces a new model, you can orient quickly: find the tier (Opus, Sonnet, or Haiku) to know where it sits on capability vs. speed, then find the class (Mythos or whatever comes next) to know which generation it belongs to.
What You Run in the App
Which Model Are You Actually Using?
When you open Claude.ai, the default selection depends on your plan. Most users interact with a Sonnet-tier model by default because it delivers a strong experience without the latency of Opus.
Claude Code fast mode is worth calling out explicitly: it uses Opus with faster output. It does not downgrade you to Sonnet or Haiku. Fast mode trades some output latency for higher throughput, but the underlying model is still Opus.
API users can pin a specific model version using its exact ID (for example, claude-sonnet-4-6) or use an alias that tracks the latest stable release in a tier. Aliases are convenient but mean your application can shift to a new version when Anthropic updates the alias target.
Common Confusion Points
Questions That Trip People Up
A few naming details consistently confuse people. Here are the most common ones.
Is Fable 5 better than Opus 4.8?
They are different things. Fable 5 is a specific model, the first public release from the Mythos class. Opus 4.8 is the flagship of the Opus tier. Which fits better depends on the task rather than a single ranking.
What does the version number mean?
It is a simple counter. Sonnet 4.6 came after Sonnet 4.5. Higher numbers within the same tier are newer and generally more capable.
Does fast mode use a smaller model?
No. Claude Code fast mode uses Opus with faster output. The model is not downgraded; throughput is optimized instead.
Why did Anthropic switch naming conventions?
The Mythos class was internal-only before Fable 5 launched in June 2026. Anthropic made the class name public when releasing Fable 5, which is why it appeared suddenly in announcements.
Go from knowing the names to using the models well
Understanding the naming system is step one. Getting good at choosing the right tier for a task, writing prompts that get the most out of each model, and building applications that use the API effectively takes practice. The Claude Academy curriculum is hands-on from the first lesson.
Frequently Asked Questions
What is the Mythos class?
Mythos is Anthropic's current model generation class. It was previously internal-only. Fable 5, released June 9, 2026, was the first Mythos model made publicly available. The class will eventually include Opus, Sonnet, and Haiku variants as Anthropic releases more models in the generation.
What is the difference between Opus, Sonnet, and Haiku?
They are tiers on a capability-versus-speed spectrum. Opus (current: 4.8) is the flagship tier with the highest capability. Sonnet (current: 4.6) is the balanced mid-tier that handles most tasks well. Haiku (current: 4.5) is the fast lightweight tier built for high-frequency, low-latency use cases. The right choice depends on what your task demands.
Is Fable 5 a replacement for Claude 3?
Fable 5 is a new-generation model from the Mythos class, which represents a step forward from prior generations. It is not a direct numbered successor in the old naming scheme. Anthropic shifted to class names (like Mythos) to distinguish major generational leaps from incremental version updates within a tier.
What model does Claude Code fast mode use?
Claude Code fast mode uses Opus with faster output. It does not switch to a smaller or lower-tier model. The tradeoff is optimized throughput, not reduced capability. This is a common misconception worth correcting: fast mode is still Opus.
How do I know which version of Claude I am talking to?
In Claude.ai, the model selector in the interface shows your current model. Via the API, the model ID in your request determines what you run (for example, claude-opus-4-8 or claude-sonnet-4-6). If you use an alias that tracks the latest stable release, check Anthropic's documentation to confirm which version the alias currently points to.
Will there be a Fable 5 Sonnet or Fable 5 Haiku?
Anthropic has not made a specific announcement, but the Mythos class is designed to span all three tiers. The pattern from prior generations suggests Sonnet and Haiku variants at the Mythos generation level are likely. Follow Anthropic's official announcements for confirmed release dates.
What does the number after the model name mean?
It is a version counter within a tier. Sonnet 4.6 is a newer update to the Sonnet tier than Sonnet 4.5 was. A higher number within the same tier generally means improved capability or reliability. The number is not a generation label; that role belongs to the class name (like Mythos).
Which Claude model should I use for coding?
For most coding tasks, Sonnet 4.6 is the practical default. It balances capability and speed well for code generation, debugging, and explanation. For complex architectural reasoning or multi-file refactors, Opus 4.8 is worth the additional latency. Haiku 4.5 suits lightweight tasks like code classification or simple completions where speed matters most.
Now put the right model to work
Knowing the names is the easy part. Using them well is the skill.