Mistral AIproprietary
Released: 2025-09-18Artificial Analysis

Magistral Medium 1.2

Magistral Medium 1.2 from Mistral, tracked on the Artificial Analysis Intelligence Index.

Intelligence Index
—/ 100
Artificial Analysis Intelligence Index (independent composite)
Throughput Speed
—tok/sec
Output streaming throughput
Time to First Answer
—ms
Time to first answer token, including any thinking
Cost to Run the Index
—USD
Artificial Analysis cost of the full Index run

Measured results · Artificial Analysis

Independent evaluations of Magistral Medium 1.2. Values are reported as published; nothing here is estimated or derived. Snapshot 2026-11-15.

Intelligence Index
—
Cost to run the Index
—
Time per task
—
Omniscience index
—
Reasoning model
Yes
Cached-input price
—

Reasoning & knowledge

Coding & terminal agents

Agentic work

Long context & documents

Multimodal & science

Full analysis on Artificial Analysis

Headline Evaluation Breakdown

Six independently measured evaluations (Artificial Analysis)
Math · AIME 202582.0%
Long context · AA-LCR53.0%
Knowledge · Humanity's Last Exam10.3%
Instruction following · IFBench43.0%

Sourced Benchmark Evaluations (12)

Verified performance across authoritative benchmarks with provenance tracking.
BenchmarkStatusScoreTrustDateSource TypeProvenance
AIMENearing Saturation82%
75
2026-11-15independentSource ↗
GPQANearing Saturation73.9%
78
2026-11-15independentSource ↗
Humanity's Last ExamActive10.3%
93
2026-11-15independentSource ↗
LiveCodeBenchSaturated75%
39
2026-11-15independentSource ↗
MMMU-ProActive59.7%
93
2026-11-15independentSource ↗
IFBenchActive43%
93
2026-11-15independentSource ↗
τ²-Bench TelecomActive52%
93
2026-11-15independentSource ↗
τ³-BankingActive5.6%
93
2026-11-15independentSource ↗
Terminal-Bench HardActive12.9%
93
2026-11-15independentSource ↗
Terminal-Bench 2.1Active12.4%
93
2026-11-15independentSource ↗
AA-LCR (Long Context Reasoning)Active53%
93
2026-11-15independentSource ↗
CritPtActive0.3%
93
2026-11-15independentSource ↗
OpenAI Compatible API ExecutionAPI Ready
from openai import OpenAI

client = OpenAI()
response = client.chat.completions.create(
    model="mistral-magistral-medium-1-2",
    messages=[{"role": "user", "content": "Evaluate multi-step logic problem"}],
    temperature=0.2
)
print(response.choices[0].message.content)

Evaluation & Sourcing Notes

Scores listed for Magistral Medium 1.2 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.