Mistral AIopen-weight
Released: 2026-04-29Artificial AnalysisMistral Medium 3.5
Mistral Medium 3.5 is a dense 128B instruction-following model from Mistral AI. It supports text and image inputs with text output, and is designed for agentic workflows, coding, and complex...
Intelligence Index
4.6/ 100
Artificial Analysis Intelligence Index (independent composite)Throughput Speed
147tok/sec
Output streaming throughputTime to First Answer
24s
Time to first answer token, including any thinkingCost to Run the Index
1.04USD
Artificial Analysis cost of the full Index runMeasured results · Artificial Analysis
Independent evaluations of Mistral Medium 3.5. Values are reported as published; nothing here is estimated or derived. Snapshot 2026-11-15.
- Intelligence Index
- 4.7
- Cost to run the Index
- $1.04
- Time per task
- 4.5 min
- Omniscience index
- -38.7
- Reasoning model
- Yes
- Cached-input price
- $0.15 / 1M
Reasoning & knowledge
- AA-Omniscience Accuracy23.0%
- CritPt0.0%
- CritPt v2 (Hard)0.5%
- GPQA Diamond74.9%
- HLE-Diamond4.7%
- Humanity's Last Exam13.8%
- IFBench68.8%
Coding & terminal agents
- FrontierCode-AA 2.01.8%
- SciCode40.2%
- Terminal-Bench 2.150.6%
- Terminal-Bench 4.00.0%
- Terminal-Bench Hard33.3%
- Terminal-Bench-Science 0.10.0%
Agentic work
- AA-AnalystAgent12.5%
- AA-Briefcase v1.1 (Elo)523 Elo
- AutomationBench-AA6.3%
- EnterpriseOps-Gym-AA33.7%
- GDPval-AA (normalized Elo)769 Elo
- τ²-Bench Telecom94.2%
- τ³-Banking15.0%
Long context & documents
Multimodal & science
- MMMU-Pro64.9%
Health & safety
Headline Evaluation Breakdown
Six independently measured evaluations (Artificial Analysis)Reasoning · CritPt v2 (Hard)0.5%
Long context · AA-LCR69.3%
Knowledge · Humanity's Last Exam13.8%
Instruction following · IFBench68.8%
Sourced Benchmark Evaluations (23)
Verified performance across authoritative benchmarks with provenance tracking.| Benchmark | Status | Score | Trust | Date | Source Type | Provenance |
|---|---|---|---|---|---|---|
| AA-LCR (Long Context Reasoning) | Active | 69.3% | 93 | 2026-11-15 | independent | Source ↗ |
| EnterpriseOps-Gym-AA | Active | 33.7% | 93 | 2026-11-15 | independent | Source ↗ |
| AA-AnalystAgent | Active | 12.5% | 93 | 2026-11-15 | independent | Source ↗ |
| AutomationBench-AA | Active | 6.3% | 93 | 2026-11-15 | independent | Source ↗ |
| MLCR-AA (Medical Long Context Reasoning) | Active | 10% | 93 | 2026-11-15 | independent | Source ↗ |
| GDP.pdf | Active | 2.8% | 93 | 2026-11-15 | independent | Source ↗ |
| FrontierCode-AA 2.0 | Active | 1.8% | 93 | 2026-11-15 | independent | Source ↗ |
| AA-Omniscience (Non-Hallucination) | Active | 19.8% | 93 | 2026-11-15 | independent | Source ↗ |
| AA-Omniscience (Accuracy) | Active | 23% | 93 | 2026-11-15 | independent | Source ↗ |
| HLE-Diamond | Active | 4.7% | 93 | 2026-11-15 | independent | Source ↗ |
| CritPt v2 (Hard) | Active | 0.5% | 93 | 2026-11-15 | independent | Source ↗ |
| CritPt | Active | 0% | 93 | 2026-11-15 | independent | Source ↗ |
| GPQA | Nearing Saturation | 74.8% | 78 | 2026-11-15 | independent | Source ↗ |
| Terminal-Bench-Science 0.1 | Active | 0% | 93 | 2026-11-15 | independent | Source ↗ |
| Terminal-Bench 4.0 | Active | 0% | 93 | 2026-11-15 | independent | Source ↗ |
| Terminal-Bench 2.1 | Active | 50.6% | 93 | 2026-11-15 | independent | Source ↗ |
| Terminal-Bench Hard | Active | 33.3% | 93 | 2026-11-15 | independent | Source ↗ |
| τ³-Banking | Active | 15.1% | 93 | 2026-11-15 | independent | Source ↗ |
| τ²-Bench Telecom | Active | 94.2% | 93 | 2026-11-15 | independent | Source ↗ |
| SciCode | Active | 40.2% | 93 | 2026-11-15 | independent | Source ↗ |
| IFBench | Active | 68.8% | 93 | 2026-11-15 | independent | Source ↗ |
| MMMU-Pro | Active | 64.9% | 93 | 2026-11-15 | independent | Source ↗ |
| Humanity's Last Exam | Active | 13.8% | 93 | 2026-11-15 | independent | Source ↗ |
Local Hardware Execution (Ollama)Hardware Compatible
# 1. Pull and execute model locally with Ollama
ollama run mistral-mistral-medium-3-5
# 2. Or invoke via local OpenAI-compatible endpoint
curl http://localhost:11434/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{
"model": "mistral-mistral-medium-3-5",
"messages": [{"role": "user", "content": "Analyze reasoning chains on AIME 2026"}]
}'Evaluation & Sourcing Notes
Scores listed for Mistral Medium 3.5 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.