GLM-4.6 (Reasoning) vs Step 3.7 Flash
Which is better for the money: GLM-4.6 (Reasoning) by Z AI or Step 3.7 Flash by StepFun? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-08-07.
Trade-off
Verdict
Depends on your budget
Step 3.7 Flash scores 1.6 points higher (30.9 vs 29.3) but costs 1.0× as much to run the benchmark ($320 vs $314; both mid-range class). Pick Step 3.7 Flash when the extra intelligence is worth the premium, otherwise GLM-4.6 (Reasoning).
| Metric | GLM-4.6 (Reasoning) | Step 3.7 Flash | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 29.3 | 30.9 | +5.5% Step 3.7 Flash is 5.5% more intelligent |
| Total AA cost (lower is better) | $314.30 | $319.97 | +1.8% Step 3.7 Flash is 1.8% more expensive |
| End-to-end time (lower is better) | 55.55s | 7.79s | −86.0% Step 3.7 Flash is 86.0% faster |
| Price per 1M tokens (in / out) | $0.55 / $2.20 | $0.20 / $1.15 | |
| Input modalities | Not reported | Not reported |
Step 3.7 Flash vs GLM-4.6 (Reasoning): 5.5% more intelligence, 1.8% more cost to run the AA benchmark, and 86.0% faster end-to-end.
Today's Pareto frontier
Against the frontier
- GLM-4.6 (Reasoning) is off the frontier: at the same or lower cost, GPT-5.6 Luna (max) scores 52.30 for $172.
- Step 3.7 Flash is off the frontier: at the same or lower cost, GPT-5.6 Luna (max) scores 52.30 for $172.
GLM-4.6 (Reasoning)Step 3.7 FlashToday'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/glm-4-6-reasoning/step-3-7-flash · Markdown: /api/v1/compare/glm-4-6-reasoning/step-3-7-flash?format=md · llms.txt · agent skill