Model comparison
Mistral Small vs Trinity Large Thinking
Trinity Large Thinking is the stronger model overall, scoring 38.6 to 33.4 on the Noometry Index.
Last verified . 20 shared benchmarks.
Mistral AI
Arcee AISummary
- They share 20 benchmarks with published results for both. Mistral Small scores higher in 1 category and Trinity Large Thinking in 7 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where Trinity Large Thinking leads 37.6 to 16.4.
- The biggest single-benchmark swing is SciCode: 26.5% for Mistral Small and 36.1% for Trinity Large Thinking.
- Mistral Small is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.25 / $0.80 for Trinity Large Thinking.
Side by side
| Mistral Small | Trinity Large Thinking | |
|---|---|---|
| Provider | Mistral AI | Arcee AI |
| Noometry Index | 33.4 | 38.6 |
| Released | 2024-02-26 | 2026-04-01 |
| Weights | Open | Open |
| Context window | 262K | 262K |
| Max output | 256K | 80K |
| Input $ / M tokens | $0.15 | $0.25 |
| Output $ / M tokens | $0.60 | $0.80 |
| Results tracked | 39 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Mistral Small: 34.0 (#247), Trinity Large Thinking: 34.1 (#244)
| Benchmark | Mistral Small | Trinity Large Thinking |
|---|---|---|
| SciCode | 26.5% | 36.1% |
| LMArena Coding | 1362 | 1381 |
| LMArena WebDev | — | 1238 |
| BigCodeBench Instruct | 36.1% | — |
| LiveBench Coding | 36.2% | — |
| BigCodeBench Complete | 46.6% | — |
| ALE-Bench | 497.62 | — |
Agentic & Tool Use Not comparable
Mistral Small: 28.1 (#93), Trinity Large Thinking: —
| Benchmark | Mistral Small | Trinity Large Thinking |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.1% | — |
Reasoning Mistral Small leads
Mistral Small: 19.8 (#250), Trinity Large Thinking: 16.9 (#298)
| Benchmark | Mistral Small | Trinity Large Thinking |
|---|---|---|
| CritPt | 0% | 0.9% |
| LMArena Hard Prompts | 1335 | 1350 |
| Kagi LLM Benchmark | 37.8% | — |
| NYT Connections (extended) | — | 16.5% |
| Thematic Generalization | — | 41.6% |
| LiveBench Reasoning | 44.8% | — |
| DTBench | 70.9% | — |
| LiveBench Data Analysis | 53.7% | — |
| LMCA | 20.6% | — |
| Surface Evolver Bench | — | 15.6% |
| LiveBench | 44% | — |
Math Trinity Large Thinking leads
Mistral Small: 16.4 (#293), Trinity Large Thinking: 37.6 (#149)
| Benchmark | Mistral Small | Trinity Large Thinking |
|---|---|---|
| LMArena Math | 1341 | 1366 |
| OTIS Mock AIME 2024-2025 | 5.8% | — |
| LiveBench Math | 39.9% | — |
| MATH Level 5 | 46.8% | — |
Knowledge Trinity Large Thinking leads
Mistral Small: 31.0 (#222), Trinity Large Thinking: 40.9 (#113)
| Benchmark | Mistral Small | Trinity Large Thinking |
|---|---|---|
| Vectara Hallucination Rate | 5.1% | 6.9% |
| LMArena Expert | 1291 | 1360 |
| GPQA Diamond | 47.5% | — |
| MMLU | 68.7% | — |
Multimodal Not comparable
Mistral Small: 33.5 (#96), Trinity Large Thinking: —
| Benchmark | Mistral Small | Trinity Large Thinking |
|---|---|---|
| LMArena Vision | 1142 | — |
Multilingual Too close to call
Mistral Small: 45.5 (#169), Trinity Large Thinking: 46.2 (#160)
| Benchmark | Mistral Small | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | 1315 | 1325 |
| LMArena Chinese | 1340 | 1373 |
| LMArena French | 1337 | 1374 |
| LMArena German | 1340 | 1356 |
| LMArena Japanese | 1275 | 1311 |
| LMArena Korean | 1259 | 1306 |
| LMArena Russian | 1324 | 1337 |
| LMArena Spanish | 1346 | 1357 |
Instruction Following Trinity Large Thinking leads
Mistral Small: 66.4 (#209), Trinity Large Thinking: 70.5 (#162)
| Benchmark | Mistral Small | Trinity Large Thinking |
|---|---|---|
| LMArena Instruction Following | 1310 | 1334 |
| LiveBench Instruction Following | 63.7% | — |
Long Context Too close to call
Mistral Small: 40.4 (#156), Trinity Large Thinking: 41.3 (#144)
| Benchmark | Mistral Small | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | 1327 | 1355 |
Writing & Preference Trinity Large Thinking leads
Mistral Small: 52.5 (#171), Trinity Large Thinking: 53.8 (#158)
| Benchmark | Mistral Small | Trinity Large Thinking |
|---|---|---|
| LMArena Text | 1338 | 1340 |
| LMArena Creative Writing | 1305 | 1320 |
| LMArena Multi-Turn | 1344 | 1342 |
| LiveBench Language | 30.5% | — |
Frequently asked questions
Is Mistral Small better than Trinity Large Thinking?
Trinity Large Thinking is the stronger model overall, scoring 38.6 to 33.4 on the Noometry Index.
Which is cheaper, Mistral Small or Trinity Large Thinking?
Mistral Small is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Trinity Large Thinking lists at $0.25 and $0.80.
Is Mistral Small or Trinity Large Thinking better for coding?
They score almost the same on coding (34.0 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 Small and Trinity Large Thinking share?
20 benchmarks have published results for both models. Mistral Small has 39 scored results on Noometry and Trinity Large Thinking has 24.