Compare AI models

Specifications
Context window
1M1M
API pricing
Input price
$0.10—
Output price
$0.50—
Cached input price
$0.01—
Cheapest input
$0.10Amazon Bedrock$0.0507Baidu
Cheapest output
$0.50Amazon Bedrock$0.2028Baidu
$45.00Anthropic list price$20.28Baidu · fp8
Benchmarks
BullshitBench v2
57%57%
Terminal-Bench 4.0
39.2%31.2%
Humanity's Last Exam · no tools
45.9%36.8%
Humanity's Last Exam · with tools
57.4%63.9%
Chartography · with tools
86.2%78.9%
threejseval
14241614
Overview
CompanyAnthropicDeepSeek
Release dateOct 7 2026Sep 10 2026
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Claude Haiku 5.5 leads DeepSeek-V4.1-Flash on 3 of the 6 benchmarks they both report. Only Claude Haiku 5.5 has a verified first-party price for an available API: $0.10 per million input tokens and $0.50 per million output tokens. No currently available pay-as-you-go API rate is tracked for DeepSeek-V4.1-Flash. DeepSeek-V4.1-Flash shipped 27 days before Claude Haiku 5.5, so benchmark comparisons should account for the intervening progress.

Context windows are 1M (Claude Haiku 5.5) vs 1M (DeepSeek-V4.1-Flash). Claude Haiku 5.5 is closed, while DeepSeek-V4.1-Flash is open weight.

On BullshitBench v2, both models score 57%. On Terminal-Bench 4.0, Claude Haiku 5.5 leads at 39.2% vs DeepSeek-V4.1-Flash at 31.2%. On Humanity's Last Exam · no tools, Claude Haiku 5.5 leads at 45.9% vs DeepSeek-V4.1-Flash at 36.8%. On Humanity's Last Exam · with tools, DeepSeek-V4.1-Flash leads at 63.9% vs Claude Haiku 5.5 at 57.4%. On Chartography · with tools, Claude Haiku 5.5 leads at 86.2% vs DeepSeek-V4.1-Flash at 78.9%. On threejseval, DeepSeek-V4.1-Flash leads at 1614 vs Claude Haiku 5.5 at 1424.