Claude 4.5 Haiku (Reasoning) vs gpt-oss-120b (High)
Which is better for the money: Claude 4.5 Haiku (Reasoning) by Anthropic or gpt-oss-120b (High) by OpenAI? 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 5.3 points higher (16.9 vs 11.6) but costs 4.8× as much to run the benchmark ($594 vs $123; premium vs mid-range cost classes). They are not alternatives at one budget: pick Claude 4.5 Haiku (Reasoning) when you need intelligence above 11.6, and gpt-oss-120b (High) when 11.6 is enough.
| Metric | Claude 4.5 Haiku (Reasoning) | gpt-oss-120b (High) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 16.9 | 11.6 | −31.4% gpt-oss-120b (High) is 31.4% less intelligent |
| Total AA cost (lower is better) | $594.08 | $122.71 | −79.3% gpt-oss-120b (High) is 79.3% cheaper |
| End-to-end time (lower is better) | 24.38s | 11.29s | −53.7% gpt-oss-120b (High) is 53.7% faster |
| Price per 1M tokens (in / out) | $1.00 / $5.00 | $0.15 / $0.60 | |
| Input modalities | Text, Image | Text |
gpt-oss-120b (High) vs Claude 4.5 Haiku (Reasoning): 31.4% less intelligence, 79.3% less cost to run the AA benchmark, and 53.7% 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.
- gpt-oss-120b (High) is off the frontier: at the same or lower cost, GPT-6 Luna (Max) scores 38.10 for $122.
Claude 4.5 Haiku (Reasoning)gpt-oss-120b (High)Today's Pareto frontierOpen-weights 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/gpt-oss-120b · Markdown: /api/v1/compare/claude-4-5-haiku-reasoning/gpt-oss-120b?format=md · llms.txt · agent skill