SQL · 5 checks · max 500 tokens · benched 2026-09-16

MiniMax M3 vs MiniMax M2.7 vs MiniMax M2.5 on monthly active users

Minimax models side by side on "Monthly active users": MiniMax M3 scores 10/10; MiniMax M3 is the cheapest answer scoring 8+ at $0.40 per 1,000 runs. Outputs, checks, judge reasons, latency and cost.

The prompt every model received

System

You are a database engineer. Reply with a single SQL statement and nothing else.

User

Table: events(user_id, event_type, occurred_at). Write one PostgreSQL query returning, for each calendar month of 2025, the month (as the first day of that month) and the number of distinct users who had at least one event in that month, ordered by month ascending.

Rubric for the judge: Uses date_trunc('month', occurred_at), COUNT(DISTINCT user_id), a 2025 filter, grouped and ordered by month.

Side by side

Every cell is one OpenRouter call at temperature 0 with the prompt's token cap and reasoning effort "low" where the model supports it. Cost is usage × the catalogue rate in models.json. Quality is one judge call to anthropic/claude-haiku-4.5 against the prompt's rubric, cached per prompt version.

MiniMax M3

minimax/minimax-m3

10/10

top judge scorecheapest scoring 8+

Latency: 1.7s
Cost: $0.00040
Per 1,000: $0.40

246 in · 294 out (214 reasoning) · 29 words · checks 5/5

regex
regex
regex
regex
contains

```sql
SELECT
    DATE_TRUNC('month', occurred_at)::date AS month,
    COUNT(DISTINCT user_id) AS distinct_users
FROM events
WHERE occurred_at >= '2025-01-01'
  AND occurred_at <  '2026-01-01'
GROUP BY DATE_TRUNC('month', occurred_at)
ORDER BY month ASC;
```

Judge: Query correctly uses date_trunc('month', occurred_at), COUNT(DISTINCT user_id), filters for 2025, groups and orders by month, and casts to date to return the first day of each month as required.

MiniMax M2.7

minimax/minimax-m2.7

7/10

Latency: 2.3s
Cost: $0.00191
Per 1,000: $1.91

90 in · 1565 out (1402 reasoning) · 38 words · checks 5/5

regex
regex
regex
regex
contains

SELECT g.month_start::date AS month,
       COUNT(DISTINCT e.user_id) AS distinct_users
FROM generate_series('2025-01-01'::timestamp,
                     '2025-12-01'::timestamp,
                     '1 month'::interval) AS g(month_start)
LEFT JOIN events e ON date_trunc('month', e.occurred_at) = g.month_start
                  AND e.occurred_at >= '2025-01-01'
                  AND e.occurred_at < '2026-01-01'
GROUP BY g.month_start
ORDER BY g.month_start;

Judge: Query correctly uses date_trunc('month'), COUNT(DISTINCT user_id), filters for 2025, groups and orders by month, and returns first day of each month; however, it uses generate_series with LEFT JOIN instead of the more direct approach of filtering events first then grouping, making it less efficient

MiniMax M2.5

minimax/minimax-m2.5

1/10

failed: empty output (finish_reason length)

Latency: 1.1s
Cost: $0.00186
Per 1,000: $1.86

90 in · 1700 out (1700 reasoning) · 0 words · checks 0/5

regex
regex
regex
regex
contains

(empty output)

Judge: No query was provided; the output is completely empty when a specific PostgreSQL query was explicitly required.

Frequently asked

What does this prompt test?

SQL: Uses date_trunc('month', occurred_at), COUNT(DISTINCT user_id), a 2025 filter, grouped and ordered by month. The deterministic checks are regex, regex, regex, regex, contains.

Which model should I pick for this task?

If the judge's bar of 8/10 is good enough for you, MiniMax M3 at $0.40 per 1,000 runs. If you need the top score, MiniMax M3 at $0.40 per 1,000 runs.

Same prompt, other families

Run it yourself

Re-run this prompt on Minimax with your own OpenRouter key, or tweak the wording and see what changes.