Claude 3.7 Sonnet
by Anthropic PBC
Retired hybrid reasoning model; use Sonnet 4.6/5 instead.
Score
Score
Our verdict
A landmark hybrid‑reasoning release that is now retired; plan migrations to Sonnet 4.6/5. Current Claude chat/Team pricing is competitive and well‑documented.
Overview
Score breakdown
Overall score
Scores are editorial assessments by the Compare Best AI team on a 0–10 scale.
Expert review
CBAI Editorial Team
Compare Best AI · Editorial Team
Claude 3.7 Sonnet marked a major step for Anthropic: a single model that could respond instantly or think in extended steps, with notably strong coding performance and a concurrent rollout of Claude Code. However, Anthropic formally retired the 3.7 Sonnet API model on Feb 19, 2026, pointing customers to Sonnet 4.6 (and, more recently, Sonnet 5) for ongoing deployments. That means you shouldn’t start new builds on 3.7 Sonnet; treat it as historical context and migrate workloads to supported Sonnet variants.
How we tested
Tasks evaluated
- ·Verified API model status and retirement dates on Anthropic’s model deprecations page.
- ·Captured current Pro/Max pricing from Claude Help Center and claude.com pricing cards.
- ·Verified Team seat prices and usage multipliers from Help Center.
- ·Verified Enterprise billing model, seat minimums, and compliance features from Help Center.
- ·Confirmed 3.7 Sonnet launch pricing/claims from Anthropic’s release post.
Method
Desk research using Anthropic/Claude official pricing pages, Help Center, and model lifecycle documentation; no hands‑on testing or performance benchmarking was performed.
Reviewer
CBAI Editorial Team
Plans & pricing
Free
Trying Claude with limited use
- Web, iOS, Android, desktop chat access
- Basic creation, coding and web search
- Memory and file creation
- Some connectors; extended thinking not included on Free
Pro
Individuals needing more usage and model access
- Everything in Free
- Includes Claude Code, Cowork, Design, Science
- Unlimited projects; Research access
- Ability to use more Claude models
- Claude for Microsoft 365
- Annual option: $17/month billed $200/year
Max 5x
Power users needing higher output limits
- ~5× Pro usage capacity per session
- Higher output limits
- Early access to advanced features
- Priority access at peak times
Max 20x
Daily heavy users needing the highest included limits
- ~20× Pro usage capacity per session
- Higher output limits
- Early access to advanced features
- Priority access at peak times
Team Standard (2–150 seats)
/seat/month
Teams needing admin controls and collaboration
- Annual option: $20/seat/month billed annually
- More usage than Pro per session
- Admin tools, billing, spend controls
- SSO, domain capture, role‑based permissions
- Connectors: Slack, Microsoft 365, Google Drive/Gmail/Calendar, GitHub
- Projects, knowledge bases, collaboration features
Team Premium (2–150 seats)
/seat/month
Power users on teams who need much higher limits
- Annual option: $100/seat/month billed annually
- Significantly more usage than Standard seats (≈6.25× Pro per session)
- All Team Standard features
Enterprise (self‑serve or sales‑assisted)
Organizations needing advanced security/compliance and usage‑metered billing
- Single seat type; billed annually; usage billed separately at API rates
- Includes Claude Code and Cowork access
- Admin cost controls; no per‑seat usage caps
- Security/compliance: audit logs, SCIM, CMEK, custom data retention, US‑only inference option, HIPAA‑readiness/BAA
- Seat minimums: 20 (self‑serve), 50 (sales‑assisted)
Pricing may vary by region. Always verify on the vendor's website.
Feature comparison
- Hybrid reasoning with standard or extended thinking; thinking budget controllable up to a 128K‑token output limit (at release).
- Strong coding/agentic workflows; release coincided with Claude Code and GitHub integration on Claude plans (at release).
- Thinking tokens were included in output‑token pricing for this model (at release).
- Multimodal capability claims at launch (text + vision inputs) with improved instruction following (at release).
- Availability at launch across Claude chat, Anthropic API, Amazon Bedrock, and Google Cloud Vertex AI; retired on Anthropic API Feb 19, 2026 (partner clouds manage their own timelines).
Is it right for you?
Good fit for
Legacy replication & audits
Researchers or teams needing to reproduce 3.7 Sonnet‑era results before migrating to Sonnet 4.6/5.
Org‑wide deployment via Claude chat
Teams standardizing on Claude chat with admin, SSO/SCIM, and compliance features (Team/Enterprise).
Less suited for
New API deployments
3.7 Sonnet is retired on the Claude API; use Sonnet 4.6/5 instead.
Those needing a fixed, unlimited API cost
Enterprise usage is metered at API rates; no unlimited token plan is offered.
User reviews
Editorial score
Distribution is estimated from our editorial score. Verified user reviews coming soon.
Use cases
Typical ways teams rely on this tool — from everyday tasks to specialized workflows.
- Legacy-to-current migration planning (3.7 Sonnet → Sonnet 4.6/5)
- Agentic coding workflows via Claude Code (successor models)
- Front‑end and full‑stack development assistance (successor models)
- Complex reasoning tasks with extended thinking (successor models)
Privacy & compliance
Security & data practices
Highlights we track from the vendor's documentation. Always confirm current terms, subprocessors, and regional policies on their official site.
- Enterprise plan includes audit logs
- SSO
- SCIM
- custom data retention controls
- Analytics & Compliance APIs
- customer‑managed encryption keys
- US‑only inference option
- and HIPAA‑readiness/BAA. Usage on Enterprise is billed at API rates on top of the annual per‑seat fee.
Details
Category
Price
- Pro $20/mo ($17/mo billed yearly). Max 5x $100/mo. Max 20x $200/mo. Team Standard $25/seat/mo ($20/seat/mo billed yearly). Team Premium $125/seat/mo ($100/seat/mo billed yearly). Enterprise: per‑user seat billed annually + usage at API rates.
Free version
Best for
- Best suited to teams needing to understand 3.7 Sonnet’s legacy capabilities and map them to active Sonnet 4.6/5 deployments
- developers working in Claude chat (Pro/Max/Team) or Enterprise with admin, SSO/SCIM, and compliance needs.
Frequently asked questions
Claude 3.7 Sonnet
Developer
Ready to get started?
Visit the Claude 3.7 Sonnet website to explore plans and start your free trial.
Was this page helpful?
Compare Best AI may earn a commission when you click links on this page. This does not influence our editorial scores or recommendations. Advertiser disclosure