Released: 2024-12-03

Amazon Nova Micro 1.0

Text-only ultra-low latency model delivering sub-second response times at an industry-disruptive price point.

Intelligence Index
73.0/ 100
Calibrated multi-domain composite
Throughput Speed
240tok/sec
Output streaming throughput
Time To First Token
140ms
Initial chunk server latency
LiveBench Contam-Free
64.2%
September 2026 suite score

Sourced Benchmark Evaluations (61)

Verified performance across authoritative benchmarks with provenance tracking.
BenchmarkStatusScoreTrustDateSource TypeProvenance
LiveBench (Continuous)Active64.2%
94
2025-02-01independentSource ↗
ARC-AGI-3Active46.2%
93
2025-01-07independentSource ↗
ARC-AGI-2Nearing Saturation55.4%
75
2024-12-31independentSource ↗
FrontierMathActive19.3%
98
2024-12-28independentSource ↗
PutnamBenchActive35.9%
90
2024-12-27independentSource ↗
Omni-MATHDeprecated51.1%
10
2024-12-25independentSource ↗
SWE-LancerDeprecated43.2%
20
2024-12-25independentSource ↗
SoohakActive52.4%
98
2024-12-23independentSource ↗
FrontierCodeActive36.8%
98
2024-12-23independentSource ↗
Humanity's Last ExamActive31.2%
93
2024-12-21independentSource ↗
StructFlowBenchNearing Saturation57.1%
75
2024-12-21independentSource ↗
AIMENearing Saturation65.6%
75
2024-12-21independentSource ↗
OlympiadBenchActive56.6%
80
2024-12-20independentSource ↗
FireBenchActive56.4%
95
2024-12-19independentSource ↗
Terminal-BenchActive44.4%
80
2024-12-19independentSource ↗
MathArenaActive55.2%
99
2024-12-19independentSource ↗
OpenAI-MRCRActive61.3%
80
2024-12-18independentSource ↗
CodeContestsDeprecated41.3%
10
2024-12-18independentSource ↗
SimpleQAActive38%
90
2024-12-18independentSource ↗
ComplexBenchNearing Saturation55.1%
80
2024-12-17independentSource ↗
SWE-bench ProActive39.4%
90
2024-12-17independentSource ↗
LongBench-ProActive59.9%
80
2024-12-17independentSource ↗
MultiPL-EDeprecated58.4%
10
2024-12-16independentSource ↗
ZeroSCROLLSNearing Saturation58.4%
78
2024-12-16independentSource ↗
AgentIFNearing Saturation45.7%
80
2024-12-15independentSource ↗
MATHSaturated67.6%
25
2024-12-15independentSource ↗
LiveCodeBenchSaturated52.1%
39
2024-12-15independentSource ↗
InfiniteBenchNearing Saturation62.1%
65
2024-12-15independentSource ↗
MMLU-ProNearing Saturation67.2%
75
2024-12-14independentSource ↗
MercuryDeprecated55.9%
10
2024-12-14independentSource ↗
SCROLLSNearing Saturation58.4%
65
2024-12-14independentSource ↗
RULERNearing Saturation67.2%
84
2024-12-13independentSource ↗
TabMWPSaturated63.5%
25
2024-12-13independentSource ↗
CyberSecEvalDeprecated57.2%
10
2024-12-13independentSource ↗
MMLU-ReduxSaturated71.5%
25
2024-12-12independentSource ↗
QuALITYNearing Saturation64.2%
78
2024-12-12independentSource ↗
SWE-bench VerifiedSaturated49.5%
25
2024-12-12independentSource ↗
MGSMNearing Saturation65.6%
65
2024-12-11independentSource ↗
LongBenchNearing Saturation62.8%
65
2024-12-11independentSource ↗
SWE-benchDeprecated43.2%
10
2024-12-11independentSource ↗
AGIEvalSaturated61.1%
25
2024-12-10independentSource ↗
QASPERNearing Saturation60.6%
65
2024-12-10independentSource ↗
TruthfulQASaturated64.2%
25
2024-12-10independentSource ↗
SuperGLUESaturated72.4%
25
2024-12-09independentSource ↗
GSM-HardDeprecated66.2%
10
2024-12-09independentSource ↗
MT-BenchNearing Saturation6.9%
75
2024-12-08independentSource ↗
BIG-Bench HardSaturated69.6%
25
2024-12-08independentSource ↗
CodeXGLUEDeprecated59.7%
23
2024-12-08independentSource ↗
MBPP+Nearing Saturation64.8%
65
2024-12-07independentSource ↗
WinoGrandeSaturated73.8%
28
2024-12-07independentSource ↗
IFEvalDeprecated65.8%
25
2024-12-07independentSource ↗
GPQANearing Saturation64.2%
78
2024-12-07independentSource ↗
HumanEval+Saturated68.6%
25
2024-12-06independentSource ↗
Needle-in-a-HaystackSaturated94.6%
44
2024-12-06independentSource ↗
CommonsenseQASaturated74.6%
28
2024-12-06independentSource ↗
GSM8KSaturated74.5%
25
2024-12-05independentSource ↗
GLUESaturated76.7%
28
2024-12-05independentSource ↗
MBPPSaturated67.3%
25
2024-12-05independentSource ↗
HumanEvalSaturated72.4%
25
2024-12-04independentSource ↗
MMLUSaturated74.5%
25
2024-12-04independentSource ↗
ARC-AGISaturated66.7%
25
2024-12-04independentSource ↗
OpenAI Compatible API ExecutionAPI Ready
from openai import OpenAI

client = OpenAI()
response = client.chat.completions.create(
    model="amazon-nova-micro-v1",
    messages=[{"role": "user", "content": "Evaluate multi-step logic problem"}],
    temperature=0.2
)
print(response.choices[0].message.content)

Evaluation & Sourcing Notes

Scores listed for Amazon Nova Micro 1.0 represent verified evaluations extracted from official research papers, independent evaluation suites (HELM, LMSYS, OpenCompass, LiveBench), and verified audit reports.

All benchmarks marked as saturated or deprecated reflect historical performance where the benchmark no longer provides active discriminative power.