Mercury 2 vs Qwen3 Coder Next
Which is better for the money: Mercury 2 by Inception or Qwen3 Coder Next by Alibaba? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-09-01.
Clear winner
Verdict
Pick Mercury 2
Qwen3 Coder Next vs Mercury 2: 2.7% less intelligence, 156.5% more cost to run the AA benchmark, and 2.7% faster end-to-end.
| Metric | Mercury 2 | Qwen3 Coder Next | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 21.9 | 21.3 | −2.7% Qwen3 Coder Next is 2.7% less intelligent |
| Total AA cost (lower is better) | $97.52 | $250.18 | +156.5% Qwen3 Coder Next is 2.6× more expensive |
| End-to-end time (lower is better) | 4.74s | 4.61s | −2.7% Qwen3 Coder Next is 2.7% faster |
| Price per 1M tokens (in / out) | $0.25 / $0.75 | $0.35 / $1.20 | |
| Input modalities | Not reported | Not reported |
Today's Pareto frontier
Against the frontier
- Mercury 2 is off the frontier: at the same or lower cost, GPT-5.6 Luna (xhigh) scores 50.10 for $96.4.
- Qwen3 Coder Next is off the frontier: at the same or lower cost, GLM-5.3-Flash scores 57.50 for $138.
Mercury 2Qwen3 Coder NextToday'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/mercury-2/qwen3-coder-next · Markdown: /api/v1/compare/mercury-2/qwen3-coder-next?format=md · llms.txt · agent skill