Claude Fable 5$22.000/MClaude Opus 4.8$11.000/MClaude Opus 4.7$11.000/MClaude Opus 4.6$11.000/MClaude Opus 4.5$33.000/MClaude Sonnet 3.7$6.600/MClaude Opus 3$33.000/MClaude 2.1$12.800/MClaude 2$12.800/MGPT-5.5$12.500/MGPT-5.2$5.425/MGPT-5.2-Codex$5.425/MGPT-5$3.875/MGPT-4.5$97.500/MGPT-4 Turbo Preview$16.000/MGPT-4$39.000/MGPT-4-32k$78.000/Mo3$19.000/Mo3-mini$2.090/Mo4-mini$2.090/Mo1$28.500/Mo1-mini$5.700/Mo1-preview$28.500/MGemini 3.5 Pro$5.000/MGemini 3.1 Pro$5.000/MGemini 3 Pro$5.000/MGemini 2.5 Pro$3.875/MGemini 1.5 Pro$2.375/MGemini 1.0 Ultra$12.000/MGemini 1.0 Pro$0.800/MClaude Fable 5$22.000/MClaude Opus 4.8$11.000/MClaude Opus 4.7$11.000/MClaude Opus 4.6$11.000/MClaude Opus 4.5$33.000/MClaude Sonnet 3.7$6.600/MClaude Opus 3$33.000/MClaude 2.1$12.800/MClaude 2$12.800/MGPT-5.5$12.500/MGPT-5.2$5.425/MGPT-5.2-Codex$5.425/MGPT-5$3.875/MGPT-4.5$97.500/MGPT-4 Turbo Preview$16.000/MGPT-4$39.000/MGPT-4-32k$78.000/Mo3$19.000/Mo3-mini$2.090/Mo4-mini$2.090/Mo1$28.500/Mo1-mini$5.700/Mo1-preview$28.500/MGemini 3.5 Pro$5.000/MGemini 3.1 Pro$5.000/MGemini 3 Pro$5.000/MGemini 2.5 Pro$3.875/MGemini 1.5 Pro$2.375/MGemini 1.0 Ultra$12.000/MGemini 1.0 Pro$0.800/M
BETA
Subscribe
← Back to Index·All Google models
Google
GoogleEfficient

Gemini 1.5 Flash-8B

Nano

Smallest Gemini 1.5 — 8B parameters, distilled for ultra-cheap inference. Punches above its size on routing/classification.

Gemini 1.5 Flash-8B is a efficient AI model from Google. It costs $0.037 per million input tokens and $0.150 per million output tokens (blended $0.071/M), with a 1M-token context window.

Released Oct 2024Modalities text, visionOfficial model page ↗Provider pricing ↗API docs ↗Compare with another model →Estimate monthly cost →
INPUT
$0.037/M
per million input tokens
OUTPUT
$0.150/M
per million output tokens
BLENDED 70/30
$0.071/M
default reference rate · how it's calculated →
CONTEXT
1M
1,000,000 tokens
What it's good at
  • Very cheap
  • Sub-100ms latency on most tokens
  • 1M context
Typical use cases
  • Routing layer
  • Bulk classification
  • Edge-style inference
Benchmarks
vs. best public score
MMLU70%
Multitask academic knowledge across 57 subjects.
GPQA Diamond38%
Graduate-level science questions, "Google-proof".
MATH58%
High-school competition math problems.
HumanEval60%
Python function synthesis from docstrings.
LMArena Elo1175 Elo
Crowd-sourced head-to-head preference Elo rating.
Hand-curated from each provider's published reports and public leaderboards. Methodology varies across sources — treat as directional rather than authoritative.
More from Google
See all 23 →
Gemini 3.5 Pro
Frontier · 1M ctx
in $2.000/Mout $12.000/M
Gemini 3.5 Flash
Balanced · 1M ctx
in $1.500/Mout $9.000/M
Gemini 3.1 Pro
Frontier · 1M ctx
in $2.000/Mout $12.000/M
Gemini 3 Pro
Frontier · 1M ctx
in $2.000/Mout $12.000/M
Gemini 3 Flash
Efficient · 1M ctx
in $0.500/Mout $3.000/M
Gemini 2.5 Pro
Frontier · 1M ctx
in $1.250/Mout $10.000/M
Frequently asked questions

How much does Gemini 1.5 Flash-8B cost?

Gemini 1.5 Flash-8B costs $0.037 per million input tokens and $0.150 per million output tokens, for a blended reference rate of $0.071 per million tokens.

What is Gemini 1.5 Flash-8B's context window?

Gemini 1.5 Flash-8B supports up to 1M tokens of context (1,000,000 tokens).

What is Gemini 1.5 Flash-8B best for?

Gemini 1.5 Flash-8B is well suited to Very cheap, Sub-100ms latency on most tokens and 1M context.

Who makes Gemini 1.5 Flash-8B?

Gemini 1.5 Flash-8B is developed and served by Google. It was released in Oct 2024.