Model comparison
Llama 3.1-8B vs Trinity Large Thinking
Trinity Large Thinking is the stronger model overall, scoring 38.6 to 23.0 on the Noometry Index. Llama 3.1-8B costs 6.7× less per token, which makes it the better buy when Trinity Large Thinking's lead doesn't matter for your workload.
Last verified . 19 shared benchmarks.
Meta
Arcee AISummary
- They share 19 benchmarks with published results for both. Llama 3.1-8B scores higher in 0 categories and Trinity Large Thinking in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Trinity Large Thinking leads 40.9 to 8.0.
- The biggest single-benchmark swing is SciCode: 13.2% for Llama 3.1-8B and 36.1% for Trinity Large Thinking.
- Llama 3.1-8B is cheaper at $0.05 / $0.08 per million input/output tokens, against $0.25 / $0.80 for Trinity Large Thinking.
- Trinity Large Thinking accepts more context: 262K tokens versus 128K.
Side by side
| Llama 3.1-8B | Trinity Large Thinking | |
|---|---|---|
| Provider | Meta | Arcee AI |
| Noometry Index | 23.0 | 38.6 |
| Released | 2024-07-23 | 2026-04-01 |
| Weights | Open | Open |
| Context window | 128K | 262K |
| Max output | 4K | 80K |
| Input $ / M tokens | $0.05 | $0.25 |
| Output $ / M tokens | $0.08 | $0.80 |
| Results tracked | 43 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Trinity Large Thinking leads
Llama 3.1-8B: 20.2 (#340), Trinity Large Thinking: 34.1 (#244)
| Benchmark | Llama 3.1-8B | Trinity Large Thinking |
|---|---|---|
| SciCode | 13.2% | 36.1% |
| LMArena Coding | 1195 | 1381 |
| LMArena WebDev | — | 1238 |
| WeirdML | 1.7% | — |
| BigCodeBench Instruct | 32.8% | — |
| BigCodeBench Complete | 40.5% | — |
| HumanEval+ | 62.8% | — |
| MBPP+ | 55.6% | — |
Agentic & Tool Use Not comparable
Llama 3.1-8B: 22.5 (#131), Trinity Large Thinking: —
| Benchmark | Llama 3.1-8B | Trinity Large Thinking |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25.8% | — |
| BALROG | 15.1% | — |
Reasoning Trinity Large Thinking leads
Llama 3.1-8B: 14.9 (#321), Trinity Large Thinking: 16.9 (#298)
| Benchmark | Llama 3.1-8B | Trinity Large Thinking |
|---|---|---|
| CritPt | 0% | 0.9% |
| LMArena Hard Prompts | 1175 | 1350 |
| NYT Connections (extended) | — | 16.5% |
| Chess Puzzles | 0% | — |
| Thematic Generalization | — | 41.6% |
| DTBench | 50.9% | — |
| LMCA | 5.4% | — |
| Surface Evolver Bench | — | 15.6% |
| Epoch Capabilities Index | 116.57 | — |
| PIQA | 81.2% | — |
Math Trinity Large Thinking leads
Llama 3.1-8B: 10.2 (#317), Trinity Large Thinking: 37.6 (#149)
| Benchmark | Llama 3.1-8B | Trinity Large Thinking |
|---|---|---|
| LMArena Math | 1179 | 1366 |
| OTIS Mock AIME 2024-2025 | 1.7% | — |
| Omni-MATH | 13.7% | — |
| MATH Level 5 | 22.9% | — |
| GSM8K | 82.4% | — |
Knowledge Trinity Large Thinking leads
Llama 3.1-8B: 8.0 (#307), Trinity Large Thinking: 40.9 (#113)
| Benchmark | Llama 3.1-8B | Trinity Large Thinking |
|---|---|---|
| LMArena Expert | 1144 | 1360 |
| GPQA Diamond | 27% | — |
| MMLU-Pro | 40.6% | — |
| Vectara Hallucination Rate | — | 6.9% |
| GPQA (HELM) | 24.7% | — |
| BoolQ | 82.8% | — |
| MMLU | 56.1% | — |
Multilingual Trinity Large Thinking leads
Llama 3.1-8B: 34.0 (#249), Trinity Large Thinking: 46.2 (#160)
| Benchmark | Llama 3.1-8B | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | 1148 | 1325 |
| LMArena Chinese | 1151 | 1373 |
| LMArena French | 1177 | 1374 |
| LMArena German | 1144 | 1356 |
| LMArena Japanese | 1061 | 1311 |
| LMArena Korean | 1053 | 1306 |
| LMArena Russian | 1158 | 1337 |
| LMArena Spanish | 1169 | 1357 |
Instruction Following Trinity Large Thinking leads
Llama 3.1-8B: 58.9 (#258), Trinity Large Thinking: 70.5 (#162)
| Benchmark | Llama 3.1-8B | Trinity Large Thinking |
|---|---|---|
| LMArena Instruction Following | 1159 | 1334 |
| IFEval | 74.3% | — |
Long Context Trinity Large Thinking leads
Llama 3.1-8B: 35.8 (#238), Trinity Large Thinking: 41.3 (#144)
| Benchmark | Llama 3.1-8B | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | 1182 | 1355 |
Writing & Preference Trinity Large Thinking leads
Llama 3.1-8B: 29.7 (#290), Trinity Large Thinking: 53.8 (#158)
| Benchmark | Llama 3.1-8B | Trinity Large Thinking |
|---|---|---|
| LMArena Text | 1187 | 1340 |
| LMArena Creative Writing | 1154 | 1320 |
| LMArena Multi-Turn | 1172 | 1342 |
| EQ-Bench Creative Writing | 713 | — |
| WildBench | 68.7% | — |
Frequently asked questions
Is Llama 3.1-8B better than Trinity Large Thinking?
Trinity Large Thinking is the stronger model overall, scoring 38.6 to 23.0 on the Noometry Index. Llama 3.1-8B costs 6.7× less per token, which makes it the better buy when Trinity Large Thinking's lead doesn't matter for your workload.
Which is cheaper, Llama 3.1-8B or Trinity Large Thinking?
Llama 3.1-8B is cheaper. It lists at $0.05 per million input tokens and $0.08 per million output tokens; Trinity Large Thinking lists at $0.25 and $0.80.
Is Llama 3.1-8B or Trinity Large Thinking better for coding?
Trinity Large Thinking scores higher on coding benchmarks: 34.1 versus 20.2 in the Noometry coding category.
Which has the bigger context window?
Trinity Large Thinking does, with 262K tokens against 128K.
How many benchmarks do Llama 3.1-8B and Trinity Large Thinking share?
19 benchmarks have published results for both models. Llama 3.1-8B has 43 scored results on Noometry and Trinity Large Thinking has 24.