Released: 2025-03-10

Amazon Nova Premier 1.0

Amazon most capable multimodal foundation model engineered for advanced enterprise agentic workflows, long-document analytics, and code generation.

Intelligence Index
90.8/ 100
Calibrated multi-domain composite
Throughput Speed
72tok/sec
Output streaming throughput
Time To First Token
390ms
Initial chunk server latency
LiveBench Contam-Free
78.2%
September 2026 suite score

Sourced Benchmark Evaluations (61)

Verified performance across authoritative benchmarks with provenance tracking.
BenchmarkStatusScoreTrustDateSource TypeProvenance
ARC-AGI-3Active58.8%
93
2025-04-14independentSource ↗
ARC-AGI-2Nearing Saturation70.6%
75
2025-04-07independentSource ↗
FrontierMathActive24.9%
98
2025-04-04independentSource ↗
PutnamBenchActive46.3%
90
2025-04-03independentSource ↗
SWE-LancerDeprecated55.4%
20
2025-04-01independentSource ↗
Omni-MATHDeprecated65.9%
10
2025-04-01independentSource ↗
SoohakActive67.6%
98
2025-03-30independentSource ↗
FrontierCodeActive47.3%
98
2025-03-30independentSource ↗
Humanity's Last ExamActive39.8%
93
2025-03-28independentSource ↗
StructFlowBenchNearing Saturation71%
75
2025-03-28independentSource ↗
AIMENearing Saturation84.6%
75
2025-03-28independentSource ↗
OlympiadBenchActive73%
80
2025-03-27independentSource ↗
MathArenaActive71.2%
99
2025-03-26independentSource ↗
FireBenchActive70.2%
95
2025-03-26independentSource ↗
Terminal-BenchActive57.1%
80
2025-03-26independentSource ↗
OpenAI-MRCRActive76.3%
80
2025-03-25independentSource ↗
SimpleQAActive47.2%
90
2025-03-25independentSource ↗
CodeContestsDeprecated53%
10
2025-03-25independentSource ↗
ComplexBenchNearing Saturation68.5%
80
2025-03-24independentSource ↗
SWE-bench ProActive50.5%
90
2025-03-24independentSource ↗
LongBench-ProActive74.5%
80
2025-03-24independentSource ↗
ZeroSCROLLSNearing Saturation72.6%
78
2025-03-23independentSource ↗
MultiPL-EDeprecated75%
10
2025-03-23independentSource ↗
LiveCodeBenchSaturated66.8%
39
2025-03-22independentSource ↗
AgentIFNearing Saturation56.8%
80
2025-03-22independentSource ↗
MATHSaturated87.2%
25
2025-03-22independentSource ↗
InfiniteBenchNearing Saturation77.2%
65
2025-03-22independentSource ↗
MercuryDeprecated71.7%
10
2025-03-21independentSource ↗
SCROLLSNearing Saturation72.6%
65
2025-03-21independentSource ↗
MMLU-ProNearing Saturation83.5%
75
2025-03-21independentSource ↗
RULERNearing Saturation83.5%
84
2025-03-20independentSource ↗
TabMWPSaturated81.9%
25
2025-03-20independentSource ↗
CyberSecEvalDeprecated73.4%
10
2025-03-20independentSource ↗
SWE-bench VerifiedSaturated63.6%
25
2025-03-19independentSource ↗
MMLU-ReduxSaturated89%
25
2025-03-19independentSource ↗
QuALITYNearing Saturation79.9%
78
2025-03-19independentSource ↗
LongBenchNearing Saturation78.1%
65
2025-03-18independentSource ↗
SWE-benchDeprecated55.4%
10
2025-03-18independentSource ↗
MGSMNearing Saturation84.6%
65
2025-03-18independentSource ↗
AGIEvalSaturated77.8%
25
2025-03-17independentSource ↗
QASPERNearing Saturation75.4%
65
2025-03-17independentSource ↗
TruthfulQASaturated79.9%
25
2025-03-17independentSource ↗
GSM-HardDeprecated85.4%
10
2025-03-16independentSource ↗
SuperGLUESaturated92.3%
25
2025-03-16independentSource ↗
CodeXGLUEDeprecated76.6%
23
2025-03-15independentSource ↗
MT-BenchNearing Saturation8.6%
75
2025-03-15independentSource ↗
BIG-Bench HardSaturated88.7%
25
2025-03-15independentSource ↗
WinoGrandeSaturated94.1%
28
2025-03-14independentSource ↗
MBPP+Nearing Saturation83.1%
65
2025-03-14independentSource ↗
IFEvalDeprecated81.9%
25
2025-03-14independentSource ↗
GPQANearing Saturation79.9%
78
2025-03-14independentSource ↗
Needle-in-a-HaystackSaturated98.2%
44
2025-03-13independentSource ↗
HumanEval+Saturated88%
25
2025-03-13independentSource ↗
CommonsenseQASaturated95%
28
2025-03-13independentSource ↗
MBPPSaturated86.4%
25
2025-03-12independentSource ↗
GSM8KSaturated96.1%
25
2025-03-12independentSource ↗
GLUESaturated97.7%
28
2025-03-12independentSource ↗
MMLUSaturated92.6%
25
2025-03-11independentSource ↗
HumanEvalSaturated92.9%
25
2025-03-11independentSource ↗
ARC-AGISaturated85.1%
25
2025-03-11independentSource ↗
LiveBench (Continuous)Active78.2%
94
2025-02-01independentSource ↗
OpenAI Compatible API ExecutionAPI Ready
from openai import OpenAI

client = OpenAI()
response = client.chat.completions.create(
    model="amazon-nova-premier-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 Premier 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.