Google: Gemini 3.5 Flash (batch)
google/gemini-3.5-flash:batch
Gemini 3.5 Flash is Google's high-efficiency multimodal model, bringing near-Pro level coding and reasoning at Flash-tier cost and speed. It is highly optimized for coding proficiency and parallel agentic execution loops, supporting text, image, video, audio, and PDF inputs. Defaults to medium thinking effort for faster and more cost-efficient responses, with full support for thinking levels (minimal, low, medium, high) for fine-grained cost/performance trade-offs.
/ 1M tokens
/ 1M tokens
tokens
Limits
Context
1,048,576
tokens
Max output
65,536
6.3% of the context window
Model facts
Author
Tokenizer
Gemini
Modality
Text + Image + File + Audio + Video -> Text
Knowledge cutoff
2025-01-01
Created
May 19, 2026
Reasoning
Publication status
Modalities
Generation controls
Capability scores
Better than 74% of models compared
Agent tasks
Category performance
1242 Elo
57.8% win
1199 Elo
57.7% win
1244 Elo
57.7% win
1244 Elo
57.5% win
1247 Elo
57.4% win
1197 Elo
56.2% win
1180 Elo
55.4% win
1194 Elo
52.6% win
1173 Elo
51.2% win
1142 Elo
46.5% win
1162 Elo
45.8% win
Additional published result
Agentic slides (HTML)
1162 Elo · Rank #6 · 45.7% win · Top 60%
1098 Elo
43.0% win
Creative tasks
Category performance
1256 Elo
58.9% win
1253 Elo
58.2% win
1245 Elo
55.8% win
1274 Elo
54.5% win
1267 Elo
54.0% win
1268 Elo
53.7% win
1236 Elo
52.9% win
1265 Elo
52.8% win