GPT-5.4 (xhigh) vs HyperNova 60B 2605
Which is better for the money: GPT-5.4 (xhigh) by OpenAI or HyperNova 60B 2605 by Multiverse Computing? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-07-17.
Trade-off
Verdict
Depends on your budget
GPT-5.4 (xhigh) scores 33.6 points higher (51.4 vs 17.8) but costs 61× as much to run the benchmark ($2132 vs $35.0; flagship vs budget cost classes). They are not alternatives at one budget: pick GPT-5.4 (xhigh) when you need intelligence above 17.8, and HyperNova 60B 2605 when 17.8 is enough.
| Metric | GPT-5.4 (xhigh) | HyperNova 60B 2605 | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 51.4 | 17.8 | −65.4% HyperNova 60B 2605 is 65.4% less intelligent |
| Total AA cost (lower is better) | $2,131.77 | $35.05 | −98.4% HyperNova 60B 2605 is 98.4% cheaper |
| End-to-end time (lower is better) | 110.80s | 8.32s | −92.5% HyperNova 60B 2605 is 92.5% faster |
| Price per 1M tokens (in / out) | $2.50 / $15.0 | $0.04 / $0.14 | |
| Input modalities | Not reported | Not reported |
HyperNova 60B 2605 vs GPT-5.4 (xhigh): 65.4% less intelligence, 98.4% less cost to run the AA benchmark, and 92.5% faster end-to-end.
Today's Pareto frontier
Against the frontier
- GPT-5.4 (xhigh) is off the frontier: at the same or lower cost, GPT-5.6 Sol (xhigh) scores 57.70 for $1543.
- HyperNova 60B 2605 is off the frontier: at the same or lower cost, MiMo-V2.5 scores 37.20 for $25.2.
GPT-5.4 (xhigh)HyperNova 60B 2605Today'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/gpt-5-4/hypernova-60b · Markdown: /api/v1/compare/gpt-5-4/hypernova-60b?format=md · llms.txt · agent skill