Claude 1
Claude 1 established a distinct frontier path emphasizing steerability, document work, and natural dialogue.
FAMILY / ANTHROPIC
Claude
A family rooted in Constitutional AI that progressed through long context, hybrid reasoning, and sustained tool use.
- First model
- 2023
- Models
- 26
- Latest dated version
- Claude Haiku 5.5
01 / RELEASE SEQUENCE
Family release sequence
- 2023-03-14Claude 1Language
- 2023-07-11Claude 2Language · Coding
- 2023-11-21Claude 2.1Language
- 2024-03-04Claude 3 OpusLanguage · Reasoning
- 2024-03-04Claude 3 SonnetLanguage · Reasoning
- 2024-03-13Claude 3 HaikuLanguage · Multimodal
- 2024-06-21Claude 3.5 SonnetLanguage · Reasoning
- 2024-10-22Claude 3.5 HaikuLanguage · Coding
- 2025-02-24Claude 3.7 SonnetLanguage · Reasoning
- 2025-05-22Claude Opus 4Reasoning · Coding
- 2025-05-22Claude Sonnet 4Reasoning · Coding
- 2025-08-05Claude Opus 4.1Language · Reasoning
- 2025-09-29Claude Sonnet 4.5Reasoning · Coding
- 2025-10-15Claude Haiku 4.5Language · Coding
- 2025-11-24Claude Opus 4.5Reasoning · Coding
- 2026-02-05Claude Opus 4.6Reasoning · Coding
- 2026-02-17Claude Sonnet 4.6Reasoning · Coding
- 2026-04-16Claude Opus 4.7Reasoning · Coding
- 2026-05-28Claude Opus 4.8Reasoning · Coding
- 2026-06-09Claude Fable 5Language · Reasoning
- 2026-06-30Claude Sonnet 5Reasoning · Coding
- 2026-07-24Claude Opus 5Reasoning · Coding
- 2026-09-01Claude Fable 5.1Language · Reasoning
- 2026-09-22Claude Opus 5.5Reasoning · Coding
- 2026-09-28Claude Sonnet 5.5Reasoning · Coding
- 2026-10-07Claude Haiku 5.5Language · Reasoning
02 / GENERATIONS
What changed
Claude 2
Claude 2 made long-document handling a central product dimension in frontier-model competition.
Claude 2.1
Expanded context to 200K, improved long-document understanding, and introduced system prompts and beta tool use.
Claude 3 Opus
Claude 3 established the Haiku/Sonnet/Opus capability tiers and made vision a standard family direction.
Claude 3 Sonnet
Sonnet established Claude’s middle capability tier and became the direct starting point for the 3.5 and 3.7 line.
Claude 3 Haiku
The fastest and lowest-cost Claude 3 model for high-throughput, low-latency workloads.
Claude 3.5 Sonnet
Claude 3.5 Sonnet became a defining 2024 general and coding model and directly preceded the hybrid-reasoning 3.7 Sonnet.
Claude 3.5 Haiku
The fast successor to Claude 3 Haiku, with stronger coding, instruction following, and tool use.
Claude 3.7 Sonnet
It merged the boundary between separate reasoners and general chat models into hybrid reasoning, while bringing coding agents into the main product line.
Claude Opus 4
The Claude 4 flagship, focused on long-running coding, complex agents, and sustained multi-hour execution.
Claude Sonnet 4
The balanced Claude 4 tier with stronger coding, reasoning, instruction following, and parallel tool use.
Claude Opus 4.1
A product upgrade to Opus 4 for coding and agent tasks, released through Claude, the API, and cloud partners.
Claude Sonnet 4.5
Strengthened long-running coding, computer use, and complex agents alongside the Claude Agent SDK.
Claude Haiku 4.5
A low-latency Haiku model offering near-Sonnet-4 coding and computer-use capability at lower cost.
Claude Opus 4.5
Introduced effort control with stronger coding, agents, computer use, and multi-agent coordination.
Claude Opus 4.6
Brought a 1M context beta to Opus with stronger long-running agents, code review, and adaptive thinking.
Claude Sonnet 4.6
A full Sonnet 4.5 upgrade across coding, computer use, long context, agent planning, and design.
Claude Opus 4.7
Focused on hard software engineering, long-run consistency, higher-resolution vision, and self-verification.
Claude Opus 4.8
Improved agent judgment, tool efficiency, computer use, and long-session collaboration over Opus 4.7.
Claude Fable 5
Fable 5 reflects a move toward layered deployment safeguards, including model fallback, rather than a single refusal policy.
Claude Sonnet 5
A new Sonnet generation with a default 1M context, adaptive thinking, and stronger agent capability closer to Opus.
Claude Opus 5
The new Opus flagship with a default 1M context and adaptive thinking for long-running agents, research, and enterprise work.
Claude Fable 5.1
Fable 5.1 makes Mythos-class capability generally available behind safeguards, combining more precise protections, lower cache-read pricing, and stronger long-horizon work in the Fable line.
Claude Opus 5.5
Opus 5.5 advances the Opus line toward more efficient long-running agent work, with lower typical workload cost and faster output than Opus 5.
Claude Sonnet 5.5
Anthropic reports 30%+ faster output and up to 30% lower per-task cost than Sonnet 5; it is the first Sonnet model with safeguards and fallbacks for high-risk cybersecurity tasks.
Claude Haiku 5.5
Haiku 5.5 advances Anthropic’s smaller Claude tier to a 1M context window and adaptive thinking. Its distinct API ID and documented specifications mark a significant new Haiku release.