Compare · ModelsLive · 2 picked · head to head
Gemini 3.5 Flash vs Grok 4.3
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Gemini 3.5 Flash wins on 15/15 benchmarks
Gemini 3.5 Flash wins 15 of 15 shared benchmarks. Leads in speed · arena · knowledge.
Category leads
speed·Gemini 3.5 Flasharena·Gemini 3.5 Flashknowledge·Gemini 3.5 Flashgeneral·Gemini 3.5 Flashmath·Gemini 3.5 Flashcoding·Gemini 3.5 Flash
Hype vs Reality
Attention vs performance
Gemini 3.5 Flash
#86 by perf·#6 by attention
Grok 4.3
#76 by perf·#16 by attention
Best value
Grok 4.3
2.9x better value than Gemini 3.5 Flash
Gemini 3.5 Flash
10.4 pts/$
$5.25/M
Grok 4.3
29.9 pts/$
$1.88/M
Vendor risk
Who is behind the model
Google DeepMind
$4.27T·Tier 1
xAI
$250.0B·Tier 1
Head to head
15 benchmarks · 2 models
Gemini 3.5 FlashGrok 4.3
Artificial Analysis · Agentic Index
Gemini 3.5 Flash leads by +13.4
Artificial Analysis Agentic Index · a composite score measuring how well a model performs in agentic workflows · multi-step tool use, planning, error recovery, and autonomous task completion. Aggregates results from multiple agentic benchmarks including SWE-bench, tool-use tests, and planning evaluations. The canonical single-number metric for "how good is this model as an agent?"
Gemini 3.5 Flash
37.5
Grok 4.3
24.1
Artificial Analysis · Coding Index
Gemini 3.5 Flash leads by +27.9
Artificial Analysis Coding Index · a composite score that aggregates performance across multiple coding benchmarks into a single index. Tracks code generation quality, debugging ability, multi-language competence, and real-world software engineering tasks. Used by Artificial Analysis to rank model coding capability in a normalized, comparable format. Useful for developers choosing between models for coding-heavy workloads.
Gemini 3.5 Flash
70.1
Grok 4.3
42.3
Artificial Analysis · Quality Index
Gemini 3.5 Flash leads by +12.6
Gemini 3.5 Flash
50.2
Grok 4.3
37.6
Chatbot Arena Elo · Coding
Gemini 3.5 Flash leads by +148.9
Gemini 3.5 Flash
1505.5
Grok 4.3
1356.6
Chatbot Arena Elo · Overall
Gemini 3.5 Flash leads by +34.8
Gemini 3.5 Flash
1476.3
Grok 4.3
1441.4
Chess Puzzles
Gemini 3.5 Flash leads by +28.4
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Gemini 3.5 Flash
42.1
Grok 4.3
13.7
Dtbench
Gemini 3.5 Flash leads by +6.7
Gemini 3.5 Flash
91.1
Grok 4.3
84.5
FrontierMath-Tier-4-v2-Private
Gemini 3.5 Flash leads by +12.2
Gemini 3.5 Flash
26.8
Grok 4.3
14.6
FrontierMath-Tiers-1-3-v2-Private
Gemini 3.5 Flash leads by +20.0
Gemini 3.5 Flash
62.8
Grok 4.3
42.8
GPQA diamond
Gemini 3.5 Flash leads by +5.3
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Gemini 3.5 Flash
90.4
Grok 4.3
85.1
Lmca
Gemini 3.5 Flash leads by +10.4
Gemini 3.5 Flash
55.5
Grok 4.3
45.1
OTIS Mock AIME 2024-2025
Gemini 3.5 Flash leads by +2.2
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Gemini 3.5 Flash
95.5
Grok 4.3
93.3
Proofbench
Gemini 3.5 Flash leads by +20.0
Gemini 3.5 Flash
31.0
Grok 4.3
11.0
SimpleQA Verified
Gemini 3.5 Flash leads by +33.0
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Gemini 3.5 Flash
66.2
Grok 4.3
33.2
WeirdML
Gemini 3.5 Flash leads by +12.8
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Gemini 3.5 Flash
62.6
Grok 4.3
49.9
Full benchmark table
| Benchmark | Gemini 3.5 Flash | Grok 4.3 |
|---|---|---|
Artificial Analysis · Agentic Index Artificial Analysis Agentic Index · a composite score measuring how well a model performs in agentic workflows · multi-step tool use, planning, error recovery, and autonomous task completion. Aggregates results from multiple agentic benchmarks including SWE-bench, tool-use tests, and planning evaluations. The canonical single-number metric for "how good is this model as an agent?" | 37.5 | 24.1 |
Artificial Analysis · Coding Index Artificial Analysis Coding Index · a composite score that aggregates performance across multiple coding benchmarks into a single index. Tracks code generation quality, debugging ability, multi-language competence, and real-world software engineering tasks. Used by Artificial Analysis to rank model coding capability in a normalized, comparable format. Useful for developers choosing between models for coding-heavy workloads. | 70.1 | 42.3 |
Artificial Analysis · Quality Index | 50.2 | 37.6 |
Chatbot Arena Elo · Coding | 1505.5 | 1356.6 |
Chatbot Arena Elo · Overall | 1476.3 | 1441.4 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 42.1 | 13.7 |
Dtbench | 91.1 | 84.5 |
FrontierMath-Tier-4-v2-Private | 26.8 | 14.6 |
FrontierMath-Tiers-1-3-v2-Private | 62.8 | 42.8 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 90.4 | 85.1 |
Lmca | 55.5 | 45.1 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 95.5 | 93.3 |
Proofbench | 31.0 | 11.0 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 66.2 | 33.2 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 62.6 | 49.9 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $1.50 | $9.00 | 1.0M tokens (~524 books) | $33.75 | |
| $1.25 | $2.50 | 1.0M tokens (~500 books) | $15.63 |
People also compared