Claude 4.5 Haiku (Reasoning) vs Gemini 3.1 Flash-Lite
Which is better for the money: Claude 4.5 Haiku (Reasoning) by Anthropic or Gemini 3.1 Flash-Lite by Google? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-26.
Trade-off
Verdict
Depends on your budget
Claude 4.5 Haiku (Reasoning) scores 1.3 points higher (16.9 vs 15.6) but costs 3.8× as much to run the benchmark ($594 vs $156; premium vs mid-range cost classes). They are not alternatives at one budget: pick Claude 4.5 Haiku (Reasoning) when you need intelligence above 15.6, and Gemini 3.1 Flash-Lite when 15.6 is enough.
| Metric | Claude 4.5 Haiku (Reasoning) | Gemini 3.1 Flash-Lite | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 16.9 | 15.6 | −7.7% Gemini 3.1 Flash-Lite is 7.7% less intelligent |
| Total AA cost (lower is better) | $594.08 | $155.87 | −73.8% Gemini 3.1 Flash-Lite is 73.8% cheaper |
| End-to-end time (lower is better) | 24.38s | 6.01s | −75.3% Gemini 3.1 Flash-Lite is 75.3% faster |
| Price per 1M tokens (in / out) | $1.00 / $5.00 | $0.25 / $1.50 | |
| Input modalities | Text, Image | Text, Image, Speech, Video |
Gemini 3.1 Flash-Lite vs Claude 4.5 Haiku (Reasoning): 7.7% less intelligence, 73.8% less cost to run the AA benchmark, and 75.3% faster end-to-end.
Today's Pareto frontier
Against the frontier
- Claude 4.5 Haiku (Reasoning) is off the frontier: at the same or lower cost, GPT-6.1 Sol (High) scores 50.20 for $521.
- Gemini 3.1 Flash-Lite is off the frontier: at the same or lower cost, GPT-6 Luna (Max) scores 38.10 for $122.
Claude 4.5 Haiku (Reasoning)Gemini 3.1 Flash-LiteToday's Pareto frontierEvery other model
Matchup
Compare something else
Agents:
Accept: text/markdown or ?format=md returns this page as Markdown. JSON: /api/v1/compare/claude-4-5-haiku-reasoning/gemini-3-1-flash-lite-preview · Markdown: /api/v1/compare/claude-4-5-haiku-reasoning/gemini-3-1-flash-lite-preview?format=md · llms.txt · agent skill