Head-to-head comparison
Claude Haiku 4.5 (Anthropic) and R1 0528 (DeepSeek) compared on the ModelCap Index, API price, context window, provider availability, weight access and every public benchmark board they share. Figures are the same ones shown on the live rankings; nothing here is a hidden score.
Snapshot as of 13 September 2026
As of 13 September 2026, R1 0528 holds the stronger ModelCap Index position (#59 vs #62); R1 0528 is 2.3× cheaper per output token ($2.15/1M vs $5.00/1M); Claude Haiku 4.5 offers the longer context window (200K vs 164K tokens); and R1 0528 ships open weights.
Evidence: 3 public benchmark observations across 3 boards · 2 public benchmark observations across 2 boards. Prices are the lowest listed API offer per million tokens observed on the OpenRouter catalogue.
Scores are the sources' own published figures for each model's best evaluated configuration; ModelCap never re-runs a benchmark.
Want a different pairing? Open the interactive comparison tool to swap either model for any current ranked language model.
R1 0528 ranks higher on the ModelCap Index as of 13 September 2026: #59 against #62. That is a capability ranking built from public benchmark evidence with published uncertainty; whether it is "better" for you also depends on price, context and where you can run it. Their published uncertainty intervals overlap, so the rank difference alone does not establish a reliable capability advantage for your workload.
R1 0528 is cheaper on output tokens: $2.15/1M against $5.00/1M. Input tokens are $1.00/1M for Claude Haiku 4.5 and $0.50/1M for R1 0528. Prices are the lowest listed API offer ModelCap observed, in USD per million tokens. For 1,000 requests with 2,000 input and 500 output tokens each (2M input + 0.5M output), the listed-rate estimate is $4.50 for Claude Haiku 4.5 versus $2.08 for R1 0528. R1 0528 costs 53.9% less in this scenario. This excludes caching, batch discounts, prompt-length tiers, tool charges and retries; verify the selected endpoint before budgeting.
Claude Haiku 4.5 has the larger context window: 200K tokens against 164K.
Arena coding: Claude Haiku 4.5 1481, R1 0528 1464 — Claude Haiku 4.5 leads.
Claude Haiku 4.5: API only. R1 0528: Open weights. Open weights mean the checkpoint can be downloaded and self-hosted under its licence; API-only models are available solely through hosted endpoints.
ModelCap currently lists 4 API providers for Claude Haiku 4.5 and 4 for R1 0528, from the OpenRouter catalogue snapshot the site serves; each model page lists the providers and their prices.
Every figure comes from the sealed ModelCap dataset published 13 September 2026; the page re-renders within a minute of each data refresh, and the ModelCap Index positions are the same ones shown on the live rankings.