Claude Haiku 5.5 (Medium) vs gpt-oss-20b (High)
Which is better for the money: Claude Haiku 5.5 (Medium) by Anthropic or gpt-oss-20b (High) by OpenAI? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-24.
Trade-off
Verdict
Depends on your budget
Claude Haiku 5.5 (Medium) scores 25.5 points higher (34.5 vs 9.0) but costs 2.0× as much to run the benchmark ($56.4 vs $28.8; mid-range vs budget cost classes). Pick Claude Haiku 5.5 (Medium) when the extra intelligence is worth the premium, otherwise gpt-oss-20b (High).
| Metric | Claude Haiku 5.5 (Medium) | gpt-oss-20b (High) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 34.5 | 9.0 | −73.9% gpt-oss-20b (High) is 73.9% less intelligent |
| Total AA cost (lower is better) | $56.36 | $28.81 | −48.9% gpt-oss-20b (High) is 48.9% cheaper |
| End-to-end time (lower is better) | 10.41s | 16.79s | +61.3% gpt-oss-20b (High) is 61.3% slower |
| Price per 1M tokens (in / out) | $0.10 / $0.50 | $0.05 / $0.22 | |
| Input modalities | Text, Image | Text |
gpt-oss-20b (High) vs Claude Haiku 5.5 (Medium): 73.9% less intelligence, 48.9% less cost to run the AA benchmark, and 61.3% slower end-to-end.
Today's Pareto frontier
Against the frontier
- Claude Haiku 5.5 (Medium) is on today's frontier: nothing is both smarter and cheaper.
- gpt-oss-20b (High) is off the frontier: at the same or lower cost, GPT-6 Luna (Low) scores 21.50 for $10.6.
Claude Haiku 5.5 (Medium)gpt-oss-20b (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-haiku-5-5-medium/gpt-oss-20b · Markdown: /api/v1/compare/claude-haiku-5-5-medium/gpt-oss-20b?format=md · llms.txt · agent skill