Model
Model explorer

Amazon Nova Micro

CLOSED
Amazon · Amazon Nova family · released Dec 3, 2024

Text-only, lowest-latency/lowest-cost Nova tier; suited to simple tasks and high-throughput agent orchestration.

ReasoningCodingVisionFunction callingTool useAgentic
923.2
Elo · rank #229
Parameters
Undisclosed
Active params
Context
128K tokens
Architecture
Transformer, architecture/size undisclosed
License
Proprietary (AWS Bedrock service terms)
Languages
200+
API price (in/out)
$0.035 / $0.14
Modalities
text
Benchmark results
Bar shows position within the tracked field; marker = field best
ARC-ChallengeReasoning90.2%#25
best: Llama 3.1 405B · 96.9%
BIG-Bench HardReasoning79.5%#34
best: ERNIE 4.5 300B-A47B · 94.3%
BFCL v3Agents56.2%#29
best: Hunyuan-A13B · 78.3%
DROPReasoning79.3#28
best: Hunyuan-T1 · 93.1
GPQA DiamondReasoning40.0%#218
best: GPT-5.6 · 94.6%
GSM8KMath92.3%#27
best: Llama 3.1 405B · 96.8%
HumanEvalCoding81.1%#52
best: Claude Opus 4.5 · 99.4%
IFEvalReasoning87.2%#43
best: Gemma 4 26B A4B · 98.5%
MATH-500Math69.3%#94
best: GPT-5 · 99.4%
MMLUKnowledge77.6%#109
best: OpenAI o3 · 92.9%
MT-BenchHuman preference8#47
best: Hunyuan-Large (A52B) · 9.4
Run it locally
Closed weights — available via API only. No local deployment.
Input / M tok
$0.035
Output / M tok
$0.14
API price $0.035/$0.14 · each benchmark row carries its own source badge (see methodology)