Tested 7 d.Revised Apr. 26
Tool review · Detailed verdict
Google
Gemini 3.1 Pro
« Long-context champion. 2M tokens is another planet entirely. »
Overall score
weighted average · 5 axes
§ Presentation video · Gemini 3.1 Pro
YouTube · officiel
§ Product preview
Price / month
19.99 €
Pro plan
Cost / 1M tokens
8 €
output rate
Context
2000k
max tokens
GDPR
6.8/10
to verify
Hosting
🇺🇸 US/UE
by default
§3.1Strength profile
Scale 0 to 10 · RadarOnAI internal method
§3.2+ Strong points
- 012M usable native context
- 02Workspace integration
- 03Native video input
§3.3− Weak points
- 01Uneven writing quality in French
- 02Product is in constant flux
Best for
Massive corpus analysis, research, video.
Not for
Polished editorial prose in French.
§04Benchmarks — Gemini 3.1 Pro
Against the rest, cold.
BenchmarkScoreValueRank
GPQA Diamond
Scientific reasoning
78.9%
#4
SWE-bench Verified
Real-world bug fixing
62%
#5
MMLU-Pro
General knowledge
85%
#1
HumanEval+
Code generation
88.1%
#4
FR-Check 2026
Internal language test
79/100
#4
Overall score evolution · Q1 '25 → Q3 '26
Q1 '25Q2 '25Q3 '25Q4 '25Q1 '26Q2 '26Q3 '26
+24
pts in 7 quarters
§ Editor's view · try it?
If your long-form text matters more than your images, Gemini 3.1 Pro remains the subscription that pays back its 19.99 € monthly fee fastest. We use it daily in the newsroom.
Affiliate link ↗ — RadarOnAI earns a commission if you subscribe, at no extra cost to you. Our verdicts never depend on commissions.
Essayer Gemini ↗
No commitment · cancel in one click