Compare · ModelsLive · 2 picked · head to head
Gemini 3.1 Pro Preview vs Gemini 3.8 Flash
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Gemini 3.8 Flash wins on 15/22 benchmarks
Gemini 3.8 Flash wins 15 of 22 shared benchmarks. Leads in speed · agentic · knowledge.
Category leads
speed·Gemini 3.8 Flashagentic·Gemini 3.8 Flashknowledge·Gemini 3.8 Flashcoding·Gemini 3.8 Flashgeneral·Gemini 3.8 Flashmath·Gemini 3.8 Flash
Hype vs Reality
Attention vs performance
Gemini 3.1 Pro Preview
#137 by perf·#5 by attention
Gemini 3.8 Flash
#63 by perf·#5 by attention
Best value
Gemini 3.8 Flash
3.7x better value than Gemini 3.1 Pro Preview
Gemini 3.1 Pro Preview
6.9 pts/$
$7.00/M
Gemini 3.8 Flash
25.5 pts/$
$2.25/M
Vendor risk
Who is behind the model
Google DeepMind
$4.20T·Tier 1
Google DeepMind
$4.20T·Tier 1
Head to head
22 benchmarks · 2 models
Gemini 3.1 Pro PreviewGemini 3.8 Flash
Artificial Analysis · CritPt
Gemini 3.8 Flash leads by +0.6
Gemini 3.1 Pro Preview
17.7
Gemini 3.8 Flash
18.3
Artificial Analysis · GDPval
Gemini 3.8 Flash leads by +31.8
Gemini 3.1 Pro Preview
13.8
Gemini 3.8 Flash
45.6
Artificial Analysis · GPQA Diamond
Gemini 3.8 Flash leads by +1.2
Gemini 3.1 Pro Preview
94.1
Gemini 3.8 Flash
95.3
Artificial Analysis · Humanity's Last Exam
Gemini 3.8 Flash leads by +0.8
Gemini 3.1 Pro Preview
47.0
Gemini 3.8 Flash
47.8
Artificial Analysis · Long Context Reasoning
Gemini 3.1 Pro Preview leads by +0.7
Gemini 3.1 Pro Preview
82.0
Gemini 3.8 Flash
81.3
Artificial Analysis · MMMU Pro
Gemini 3.8 Flash leads by +3.2
Gemini 3.1 Pro Preview
82.4
Gemini 3.8 Flash
85.6
Artificial Analysis · Quality Index
Gemini 3.8 Flash leads by +11.2
Gemini 3.1 Pro Preview
29.7
Gemini 3.8 Flash
40.9
Artificial Analysis · SciCode
Gemini 3.1 Pro Preview leads by +2.1
Gemini 3.1 Pro Preview
58.7
Gemini 3.8 Flash
56.6
APEX-Agents
Gemini 3.8 Flash leads by +29.0
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Gemini 3.1 Pro Preview
35.3
Gemini 3.8 Flash
64.3
Chess Puzzles
Gemini 3.8 Flash leads by +6.3
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Gemini 3.1 Pro Preview
52.6
Gemini 3.8 Flash
59.0
Deepswe
Gemini 3.8 Flash leads by +62.1
Gemini 3.1 Pro Preview
11.7
Gemini 3.8 Flash
73.8
Dtbench
Gemini 3.1 Pro Preview leads by +2.3
Gemini 3.1 Pro Preview
95.1
Gemini 3.8 Flash
92.9
FrontierMath-Tier-4-v2-Private
Gemini 3.1 Pro Preview leads by +4.9
Gemini 3.1 Pro Preview
26.8
Gemini 3.8 Flash
21.9
FrontierMath-Tiers-1-3-v2-Private
Gemini 3.8 Flash leads by +8.8
Gemini 3.1 Pro Preview
59.6
Gemini 3.8 Flash
68.4
Furniture Assembly
Gemini 3.8 Flash leads by +2.4
Gemini 3.1 Pro Preview
0.0
Gemini 3.8 Flash
2.4
GPQA diamond
Gemini 3.8 Flash leads by +1.3
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Gemini 3.1 Pro Preview
92.6
Gemini 3.8 Flash
93.9
HLE
Gemini 3.1 Pro Preview leads by +2.0
HLE (Humanity's Last Exam) · a reasoning benchmark designed to be the hardest public evaluation of AI. Questions span mathematics, physics, philosophy, and logic · curated to be at or beyond the frontier of human expert capability. Tested with and without tool augmentation. Claude Opus 4.7 scores 46.9% without tools and 54.7% with tools · making it one of the few benchmarks where the top score is below 60%.
Gemini 3.1 Pro Preview
43.7
Gemini 3.8 Flash
41.7
Lmca
Gemini 3.1 Pro Preview leads by +1.1
Gemini 3.1 Pro Preview
63.3
Gemini 3.8 Flash
62.2
Mystery Game Puzzles
Gemini 3.8 Flash leads by +14.3
Gemini 3.1 Pro Preview
27.3
Gemini 3.8 Flash
41.6
OTIS Mock AIME 2024-2025
Gemini 3.8 Flash leads by +3.3
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Gemini 3.1 Pro Preview
95.6
Gemini 3.8 Flash
98.9
Proofbench
Gemini 3.8 Flash leads by +22.0
Gemini 3.1 Pro Preview
26.0
Gemini 3.8 Flash
48.0
SimpleQA Verified
Gemini 3.1 Pro Preview leads by +3.8
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Gemini 3.1 Pro Preview
73.5
Gemini 3.8 Flash
69.7
Full benchmark table
| Benchmark | Gemini 3.1 Pro Preview | Gemini 3.8 Flash |
|---|---|---|
Artificial Analysis · CritPt | 17.7 | 18.3 |
Artificial Analysis · GDPval | 13.8 | 45.6 |
Artificial Analysis · GPQA Diamond | 94.1 | 95.3 |
Artificial Analysis · Humanity's Last Exam | 47.0 | 47.8 |
Artificial Analysis · Long Context Reasoning | 82.0 | 81.3 |
Artificial Analysis · MMMU Pro | 82.4 | 85.6 |
Artificial Analysis · Quality Index | 29.7 | 40.9 |
Artificial Analysis · SciCode | 58.7 | 56.6 |
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 35.3 | 64.3 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 52.6 | 59.0 |
Deepswe | 11.7 | 73.8 |
Dtbench | 95.1 | 92.9 |
FrontierMath-Tier-4-v2-Private | 26.8 | 21.9 |
FrontierMath-Tiers-1-3-v2-Private | 59.6 | 68.4 |
Furniture Assembly | 0.0 | 2.4 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 92.6 | 93.9 |
HLE HLE (Humanity's Last Exam) · a reasoning benchmark designed to be the hardest public evaluation of AI. Questions span mathematics, physics, philosophy, and logic · curated to be at or beyond the frontier of human expert capability. Tested with and without tool augmentation. Claude Opus 4.7 scores 46.9% without tools and 54.7% with tools · making it one of the few benchmarks where the top score is below 60%. | 43.7 | 41.7 |
Lmca | 63.3 | 62.2 |
Mystery Game Puzzles | 27.3 | 41.6 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 95.6 | 98.9 |
Proofbench | 26.0 | 48.0 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 73.5 | 69.7 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $2.00 | $12.00 | 1.0M tokens (~524 books) | $45.00 | |
| $0.75 | $3.75 | 1.0M tokens (~524 books) | $15.00 |