Model comparison
Mistral Medium vs Trinity Large Thinking
Trinity Large Thinking is the stronger model overall, scoring 38.6 to 36.3 on the Noometry Index.
Last verified . 21 shared benchmarks.
Mistral AI
Arcee AISummary
- They share 21 benchmarks with published results for both. Mistral Medium scores higher in 6 categories and Trinity Large Thinking in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Trinity Large Thinking leads 40.9 to 25.0.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 22.7% for Mistral Medium and 6.9% for Trinity Large Thinking.
- Trinity Large Thinking is cheaper at $0.25 / $0.80 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
Side by side
| Mistral Medium | Trinity Large Thinking | |
|---|---|---|
| Provider | Mistral AI | Arcee AI |
| Noometry Index | 36.3 | 38.6 |
| Released | 2023-12-11 | 2026-04-01 |
| Weights | Open | Open |
| Context window | 262K | 262K |
| Max output | 262K | 80K |
| Input $ / M tokens | $1.50 | $0.25 |
| Output $ / M tokens | $7.50 | $0.80 |
| Results tracked | 36 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Mistral Medium: 34.2 (#243), Trinity Large Thinking: 34.1 (#244)
| Benchmark | Mistral Medium | Trinity Large Thinking |
|---|---|---|
| SciCode | 40.2% | 36.1% |
| LMArena Coding | 1434 | 1381 |
| FrontierCode | 8% | — |
| LMArena WebDev | — | 1238 |
| WeirdML | 43.7% | — |
| ALE-Bench | 763.98 | — |
Agentic & Tool Use Not comparable
Mistral Medium: 28.3 (#90), Trinity Large Thinking: —
| Benchmark | Mistral Medium | Trinity Large Thinking |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.7% | — |
Reasoning Mistral Medium leads
Mistral Medium: 24.0 (#167), Trinity Large Thinking: 16.9 (#298)
| Benchmark | Mistral Medium | Trinity Large Thinking |
|---|---|---|
| CritPt | 0% | 0.9% |
| LMArena Hard Prompts | 1426 | 1350 |
| Surface Evolver Bench | 26.9% | 15.6% |
| Kagi LLM Benchmark | 50% | — |
| NYT Connections (extended) | — | 16.5% |
| Thematic Generalization | — | 41.6% |
| DTBench | 75.5% | — |
| LMCA | 26.1% | — |
Math Trinity Large Thinking leads
Mistral Medium: 28.1 (#245), Trinity Large Thinking: 37.6 (#149)
| Benchmark | Mistral Medium | Trinity Large Thinking |
|---|---|---|
| LMArena Math | 1408 | 1366 |
| OTIS Mock AIME 2024-2025 | 32.2% | — |
| ProofBench | 9% | — |
| MATH Level 5 | 81.6% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Trinity Large Thinking leads
Mistral Medium: 25.0 (#265), Trinity Large Thinking: 40.9 (#113)
| Benchmark | Mistral Medium | Trinity Large Thinking |
|---|---|---|
| Vectara Hallucination Rate | 22.7% | 6.9% |
| LMArena Expert | 1408 | 1360 |
| GPQA Diamond | 59.5% | — |
| Humanity's Last Exam | 4.5% | — |
Multimodal Not comparable
Mistral Medium: 35.3 (#88), Trinity Large Thinking: —
| Benchmark | Mistral Medium | Trinity Large Thinking |
|---|---|---|
| LMArena Vision | 1172 | — |
Multilingual Mistral Medium leads
Mistral Medium: 52.1 (#91), Trinity Large Thinking: 46.2 (#160)
| Benchmark | Mistral Medium | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | 1408 | 1325 |
| LMArena Chinese | 1447 | 1373 |
| LMArena French | 1459 | 1374 |
| LMArena German | 1432 | 1356 |
| LMArena Japanese | 1378 | 1311 |
| LMArena Korean | 1380 | 1306 |
| LMArena Russian | 1411 | 1337 |
| LMArena Spanish | 1433 | 1357 |
Instruction Following Mistral Medium leads
Mistral Medium: 73.7 (#116), Trinity Large Thinking: 70.5 (#162)
| Benchmark | Mistral Medium | Trinity Large Thinking |
|---|---|---|
| LMArena Instruction Following | 1398 | 1334 |
Long Context Mistral Medium leads
Mistral Medium: 42.9 (#114), Trinity Large Thinking: 41.3 (#144)
| Benchmark | Mistral Medium | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | 1406 | 1355 |
Writing & Preference Mistral Medium leads
Mistral Medium: 60.0 (#103), Trinity Large Thinking: 53.8 (#158)
| Benchmark | Mistral Medium | Trinity Large Thinking |
|---|---|---|
| LMArena Text | 1424 | 1340 |
| LMArena Creative Writing | 1391 | 1320 |
| LMArena Multi-Turn | 1418 | 1342 |
| Short-Story Creative Writing | 77.3% | — |
Frequently asked questions
Is Mistral Medium better than Trinity Large Thinking?
Trinity Large Thinking is the stronger model overall, scoring 38.6 to 36.3 on the Noometry Index.
Which is cheaper, Mistral Medium or Trinity Large Thinking?
Trinity Large Thinking is cheaper. It lists at $0.25 per million input tokens and $0.80 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is Mistral Medium or Trinity Large Thinking better for coding?
They score almost the same on coding (34.2 vs 34.1); test both on your own repository before choosing.
Which has the bigger context window?
Both accept 262K tokens.
How many benchmarks do Mistral Medium and Trinity Large Thinking share?
21 benchmarks have published results for both models. Mistral Medium has 36 scored results on Noometry and Trinity Large Thinking has 24.