Coding · Google DeepMind

Gemini 3 Flash Vision on LiveCodeBench: score, rank and what it means

Gemini 3 Flash Vision scores 73.2 on LiveCodeBench, placing it #122 of 600 models in the index. The benchmark average is 61.0 and the leader is Claude Sonnet 4.5 Reasoning at 85.1.

Score
73.2
0–100 scale
Rank
#122
of 600
Percentile
80th
within the index
Benchmark avg
61.0
all models
Leader
85.1
Claude Sonnet 4.5 Reasoning
Gap to leader
11.9
points

Score distribution

The chart shows the top 10 models on LiveCodeBench, with Gemini 3 Flash Vision highlighted. Longer bars mean higher scores.

Claude Sonnet 4.5 Reasoning 85.1
Grok 4 Reasoning 85.1
Claude Opus 4.5 85.0
GPT-5.2 Max 84.8
Grok 4 Turbo 84.7
GPT-5.1 Turbo 84.6
GPT-5.2 Pro 84.6
GPT-5.2 Reasoning 84.6
Claude Opus 4.5 Pro 84.1
Gemini 3 Pro Max 83.6

Top 10 leaderboard

#ModelProviderContextInput / 1MAvg score
1 Claude Sonnet 4.5 Reasoning Anthropic 400000 $19.48 87.7
2 Grok 4 Reasoning xAI 2000000 $19.67 87.8
3 Claude Opus 4.5 Anthropic 400000 $19.42 87.7
4 GPT-5.2 Max OpenAI 1000000 $21.39 89.2
5 Grok 4 Turbo xAI 1000000 $19.21 86.7
6 GPT-5.1 Turbo OpenAI 2000000 $19.19 86.1
7 GPT-5.2 Pro OpenAI 2000000 $21.32 89.6
8 GPT-5.2 Reasoning OpenAI 1000000 $20.98 87.5
9 Claude Opus 4.5 Pro Anthropic 200000 $20.98 89.1
10 Gemini 3 Pro Max Google DeepMind 2000000 $20.57 87.9

What this score means

On LiveCodeBench, Gemini 3 Flash Vision sits at #122 of 600. The leader, Claude Sonnet 4.5 Reasoning, scores 85.1, so the gap to the top is 11.9 points. The median model in this index scores 64.7, which puts Gemini 3 Flash Vision above the midpoint.

Google DeepMind models average 70.8 on this benchmark. Gemini 3 Flash Vision is 2.4 points above that provider line, so it is a stronger result than the average Google DeepMind model.

Among multimodal models, the average is 70.3. Gemini 3 Flash Vision is 2.9 points above that group. This is the more useful comparison if you are shortlisting within the same tier rather than across the whole index.

Read a 73.2 as a relative signal, not an absolute guarantee. Benchmark scores depend on prompt format, evaluation harness and contamination controls. Use this page to narrow the shortlist, then test the model on your own data.

Price context

At $15.06 per million input tokens, Gemini 3 Flash Vision is above the median input price of $4.00 across the 600 models scored on LiveCodeBench. The models immediately around it in the ranking are shown below so you can compare score and price together.

ModelProviderContextInput / 1MAvg score
Magistral Instruct Mistral AI 65536 $4.48 78.4
Doubao 1.5 Pro Max ByteDance 400000 $14.93 81.2
Nova Pro Amazon 32768 $12.75 79.4
MiniMax-M2 MiniMax 200000 $5.19 80.6
Gemini 3 Flash Vision Google DeepMind 400000 $15.06 81.9
MiniMax-Text Chat MiniMax 128000 $4.26 77.8
Command A Max Cohere 400000 $14.28 79.9

Other benchmarks for Gemini 3 Flash Vision

BenchmarkScoreRankPercentileRelative
MMLU-Pro 81.8 125 79th
GPQA Diamond 75.7 130 78th
SWE-bench Verified 71.4 86 86th
HumanEval 84.5 124 79th
MATH-500 87.5 72 88th
AIME 2025 82.4 61 90th
IFEval 86.9 135 78th
TruthfulQA 85.5 59 90th
HellaSwag 91.5 47 92th
MMMU 80.3 55 91th
Arena Elo 1430.3 120 80th

FAQ

What score does Gemini 3 Flash Vision get on LiveCodeBench?

Gemini 3 Flash Vision scores 73.2 on LiveCodeBench in this index. The benchmark average is 61.0 and the top score is 85.1.

How does Gemini 3 Flash Vision rank on LiveCodeBench?

It ranks #122 of 600 models, in the 80th percentile. The leader is Claude Sonnet 4.5 Reasoning at 85.1.

Is Gemini 3 Flash Vision good at LiveCodeBench?

Yes, relative to the index: it is above the median and inside the top half of scores. The result should be read alongside its price and context window.

How is the LiveCodeBench score calculated?

LiveCodeBench is a coding benchmark. Scores in this index are modelled estimates on a 0–100 scale (or Elo for Arena), derived from capability tier and benchmark difficulty. They are not official leaderboard results.

What does Gemini 3 Flash Vision cost?

Gemini 3 Flash Vision costs $15.06 per million input tokens and $82.14 per million output tokens.

Related