MiniMaxproprietary
Released: 2026-06-25

MiniMax M3

MiniMax frontier model with MiniMax Sparse Attention (MSA), 1M context, and top SWE-Bench Pro agentic completion.

Intelligence Index
89.5/ 100
Calibrated multi-domain composite
Throughput Speed
88tok/sec
Output streaming throughput
Time To First Token
340ms
Initial chunk server latency
LiveBench Contam-Free
76.8%
September 2026 suite score

Sourced Benchmark Evaluations (2)

Verified performance across authoritative benchmarks with provenance tracking.
BenchmarkStatusScoreTrustDateSource TypeProvenance
ARC-AGI-2Nearing Saturation68.6%
75
2026-07-23independentSource ↗
SWE-bench VerifiedSaturated60.8%
25
2026-07-04independentSource ↗
OpenAI Compatible API ExecutionAPI Ready
from openai import OpenAI

client = OpenAI()
response = client.chat.completions.create(
    model="minimax-m3",
    messages=[{"role": "user", "content": "Evaluate multi-step logic problem"}],
    temperature=0.2
)
print(response.choices[0].message.content)

Evaluation & Sourcing Notes

Scores listed for MiniMax M3 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.