Model comparison
Mistral Large vs Trinity Large Thinking
Trinity Large Thinking is the stronger model overall, scoring 38.6 to 31.9 on the Noometry Index.
Last verified . 20 shared benchmarks.
Mistral AI
Arcee AISummary
- They share 20 benchmarks with published results for both. Mistral Large scores higher in 1 category and Trinity Large Thinking in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Trinity Large Thinking leads 37.6 to 18.2.
- Trinity Large Thinking is cheaper at $0.25 / $0.80 per million input/output tokens, against $2 / $6 for Mistral Large.
- Trinity Large Thinking accepts more context: 262K tokens versus 131K.
Side by side
| Mistral Large | Trinity Large Thinking | |
|---|---|---|
| Provider | Mistral AI | Arcee AI |
| Noometry Index | 31.9 | 38.6 |
| Released | 2024-02-26 | 2026-04-01 |
| Weights | Open | Open |
| Context window | 131K | 262K |
| Max output | 16K | 80K |
| Input $ / M tokens | $2 | $0.25 |
| Output $ / M tokens | $6 | $0.80 |
| Results tracked | 51 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Mistral Large: 34.3 (#240), Trinity Large Thinking: 34.1 (#244)
| Benchmark | Mistral Large | Trinity Large Thinking |
|---|---|---|
| SciCode | 36.2% | 36.1% |
| LMArena Coding | 1277 | 1381 |
| LMArena WebDev | — | 1238 |
| BigCodeBench Instruct | 30% | — |
| LiveBench Coding | 47.1% | — |
| BigCodeBench Complete | 38.3% | — |
| ALE-Bench | 264.7 | — |
| HumanEval+ | 62.2% | — |
| MBPP+ | 59.5% | — |
Agentic & Tool Use Not comparable
Mistral Large: 28.6 (#89), Trinity Large Thinking: —
| Benchmark | Mistral Large | Trinity Large Thinking |
|---|---|---|
| Berkeley Function Calling Leaderboard | 38.4% | — |
Reasoning Trinity Large Thinking leads
Mistral Large: 15.8 (#310), Trinity Large Thinking: 16.9 (#298)
| Benchmark | Mistral Large | Trinity Large Thinking |
|---|---|---|
| CritPt | 0% | 0.9% |
| LMArena Hard Prompts | 1257 | 1350 |
| SimpleBench | 22.5% | — |
| NYT Connections (extended) | — | 16.5% |
| Thematic Generalization | — | 41.6% |
| LiveBench Reasoning | 43.5% | — |
| DTBench | 65.1% | — |
| LiveBench Data Analysis | 50.1% | — |
| LMCA | 16.7% | — |
| Surface Evolver Bench | — | 15.6% |
| Epoch Capabilities Index | 128.52 | — |
| ForecastBench | 57.1 | — |
| LiveBench | 48.4% | — |
Math Trinity Large Thinking leads
Mistral Large: 18.2 (#291), Trinity Large Thinking: 37.6 (#149)
| Benchmark | Mistral Large | Trinity Large Thinking |
|---|---|---|
| LMArena Math | 1262 | 1366 |
| OTIS Mock AIME 2024-2025 | 8.5% | — |
| Omni-MATH | 28.1% | — |
| LiveBench Math | 42.5% | — |
| MATH Level 5 | 50.3% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Trinity Large Thinking leads
Mistral Large: 30.1 (#230), Trinity Large Thinking: 40.9 (#113)
| Benchmark | Mistral Large | Trinity Large Thinking |
|---|---|---|
| Vectara Hallucination Rate | 4.5% | 6.9% |
| LMArena Expert | 1232 | 1360 |
| GPQA Diamond | 51.3% | — |
| MMLU-Pro | 59.9% | — |
| Confabulations | 21.4% | — |
| GPQA (HELM) | 43.5% | — |
| MMLU | 80% | — |
Multilingual Trinity Large Thinking leads
Mistral Large: 40.0 (#219), Trinity Large Thinking: 46.2 (#160)
| Benchmark | Mistral Large | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | 1237 | 1325 |
| LMArena Chinese | 1240 | 1373 |
| LMArena French | 1325 | 1374 |
| LMArena German | 1254 | 1356 |
| LMArena Japanese | 1188 | 1311 |
| LMArena Korean | 1202 | 1306 |
| LMArena Russian | 1257 | 1337 |
| LMArena Spanish | 1268 | 1357 |
Instruction Following Trinity Large Thinking leads
Mistral Large: 67.9 (#191), Trinity Large Thinking: 70.5 (#162)
| Benchmark | Mistral Large | Trinity Large Thinking |
|---|---|---|
| LMArena Instruction Following | 1249 | 1334 |
| LiveBench Instruction Following | 67.9% | — |
| IFEval | 87.7% | — |
Long Context Trinity Large Thinking leads
Mistral Large: 38.3 (#199), Trinity Large Thinking: 41.3 (#144)
| Benchmark | Mistral Large | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | 1261 | 1355 |
Writing & Preference Trinity Large Thinking leads
Mistral Large: 40.7 (#242), Trinity Large Thinking: 53.8 (#158)
| Benchmark | Mistral Large | Trinity Large Thinking |
|---|---|---|
| LMArena Text | 1266 | 1340 |
| LMArena Creative Writing | 1243 | 1320 |
| LMArena Multi-Turn | 1260 | 1342 |
| Short-Story Creative Writing | 69% | — |
| EQ-Bench Creative Writing | 985 | — |
| WildBench | 80.1% | — |
| LiveBench Language | 39.4% | — |
Frequently asked questions
Is Mistral Large better than Trinity Large Thinking?
Trinity Large Thinking is the stronger model overall, scoring 38.6 to 31.9 on the Noometry Index.
Which is cheaper, Mistral Large 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 Large lists at $2 and $6.
Is Mistral Large or Trinity Large Thinking better for coding?
They score almost the same on coding (34.3 vs 34.1); test both on your own repository before choosing.
Which has the bigger context window?
Trinity Large Thinking does, with 262K tokens against 131K.
How many benchmarks do Mistral Large and Trinity Large Thinking share?
20 benchmarks have published results for both models. Mistral Large has 51 scored results on Noometry and Trinity Large Thinking has 24.