Model
Model explorer

Gemini 1.5 Flash

CLOSED
Google · Gemini 1.5 family · released May 14, 2024

Distilled fast/cheap 1.5-series workhorse that became the free-tier Gemini model; 1M-token multimodal context.

ReasoningCodingVisionFunction callingTool useAgentic
865.5
Elo · rank #244
Parameters
Undisclosed
Active params
—
Context
1M tokens
Architecture
Dense Transformer decoder model (parameter count undisclosed), online-distilled from the larger Gemini 1.5 Pro
License
Proprietary
Languages
100+
API price (in/out)
$0.075 / $0.3
Modalities
text · vision · audio · video
Benchmark results
Bar shows position within the tracked field; marker = field best
AI2DVision91.7%#18
best: Molmo 72B · 96.3%
AMCMath34.8%#6
best: Falcon-H1 34B · 69.4%
BIG-Bench HardReasoning85.5%#16
best: ERNIE 4.5 300B-A47B · 94.3%
BFCLAgents81.8%#8
best: SmolLM3 3B (No Thinking) · 92.3%
ChartQAVision85.4%#23
best: MiniMax-VL-01 · 91.7%
DocVQAVision89.9%#33
best: Qwen2-VL-72B · 96.5%
DROPReasoning78.4#32
best: Hunyuan-T1 · 93.1
GPQA DiamondReasoning39.5%#222
best: GPT-5.6 · 94.6%
GSM8KMath86.2%#60
best: Llama 3.1 405B · 96.8%
HellaSwagReasoning86.5%#24
best: Claude 3 Opus · 95.4%
HumanEvalCoding74.3%#68
best: Claude Opus 4.5 · 99.4%
LogicKorHuman preference8.77#9
best: GPT-4o · 9.33
MATH-500Math54.9%#119
best: GPT-5 · 99.4%
MathVistaVision58.4%#45
best: Seed 2.1 Pro · 90.7%
MGSMMath82.6%#25
best: OpenAI o4-mini · 93.7%
MMLUKnowledge78.9%#99
best: OpenAI o3 · 92.9%
MMMUVision56.1%#83
best: Claude Fable 5 · 89.3%
TextVQAVision78.7%#14
best: Molmo 2 8B · 85.7%
VQAv2Vision80.1%#8
best: Reka Edge (2026) · 88.4%
Run it locally
Closed weights — available via API only. No local deployment.
Input / M tok
$0.075
Output / M tok
$0.3
Gemini 1.5 family
Elo progression across releases
Gemini 1.5 ProFeb 20241062.3Gemini 1.5 FlashMay 2024865.5Gemini 1.5 Flash-8BOct 2024532.6
Resources
DOCSai.google.dev/gemini-api/docs/modelsAPIAvailable via Google API
API price $0.075/$0.3 · each benchmark row carries its own source badge (see methodology)