Claude Haiku 4.5: added Anthropic's published SWE-bench Verified result of 73.3%
Anthropic's latest model achieves a record-breaking benchmark score for efficient code generation.
Added Anthropic's published SWE-bench Verified result of 73.3% — verified by the OnAirToday AI Progress Tracker against Anthropic's published documentation.
OnAirToday
Newsroom
The OnAirToday AI Progress Tracker recorded a verified change to Anthropic's Claude Haiku 4.5 (version claude-haiku-4-5): Added Anthropic's published SWE-bench Verified result of 73.3%.
Claude Haiku 4.5 is Anthropic's speed-focused model for coding, conversational applications and tool-driven workflows, with support for extended thinking and computer use. It has a 200K-token context window and standard API pricing of $1 per million input tokens and $5 per million output tokens.
What changed in this review:
- Documented extended thinking, computer use and improvements over earlier Haiku models.
- Capability score 49 → 73
- Flaws and limitations updated
- Risks and threats updated
Teams building on Claude Haiku 4.5 should confirm limits, pricing and benchmark coverage against the vendor's own documentation before relying on these figures.
The full profile — capabilities, benchmark citations, known flaws and risk guidance — is on the tracker at /ai-models/claude-haiku-4-5.
References:
- https://www.anthropic.com/news/claude-haiku-4-5
- https://docs.anthropic.com/en/docs/about-claude/models/overview
Generated insight by OnAirToday AI
Claude Haiku 4.5 achieving 73.3% on SWE-bench Verified marks a significant leap in efficiency and coding capabilities for lightweight models. This performance level demonstrates that smaller architectures are increasingly capable of solving complex, real-world software engineering tasks previously reserved for massive LLMs.

Comments
Keep comments relevant and respectful. No links, advertising, self-promotion, or spam.