Model comparison
Inkling-Small vs Llama 3.1-405B
Inkling-Small is the stronger model overall, scoring 46.5 to 30.7 on the Noometry Index.
Last verified . 21 shared benchmarks.
MetaSummary
- They share 21 benchmarks with published results for both. Inkling-Small scores higher in 8 categories and Llama 3.1-405B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Inkling-Small leads 45.1 to 18.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 90% for Inkling-Small and 9.7% for Llama 3.1-405B.
Side by side
| Inkling-Small | Llama 3.1-405B | |
|---|---|---|
| Provider | Thinking Machines Lab | Meta |
| Noometry Index | 46.5 | 30.7 |
| Released | 2026-07-15 | 2024-07-23 |
| Weights | Open | Open |
| Context window | 524K | — |
| Max output | 1.05M | — |
| Input $ / M tokens | $0.45 | — |
| Output $ / M tokens | $1.20 | — |
| Results tracked | 33 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Inkling-Small leads
Inkling-Small: 43.6 (#85), Llama 3.1-405B: 33.1 (#262)
| Benchmark | Inkling-Small | Llama 3.1-405B |
|---|---|---|
| LMArena Coding | 1451 | 1291 |
| LMArena WebDev | 1409 | — |
| SciCode | 48.7% | — |
| WeirdML | — | 21.4% |
Agentic & Tool Use Not comparable
Inkling-Small: —, Llama 3.1-405B: 21.0 (#140)
| Benchmark | Inkling-Small | Llama 3.1-405B |
|---|---|---|
| TheAgentCompany | — | 7.4% |
| Cybench | — | 7.5% |
Reasoning Inkling-Small leads
Inkling-Small: 38.6 (#63), Llama 3.1-405B: 16.8 (#300)
| Benchmark | Inkling-Small | Llama 3.1-405B |
|---|---|---|
| LMArena Hard Prompts | 1423 | 1269 |
| Epoch Capabilities Index | 150.15 | 128.75 |
| ARC-AGI-2 | 40.1% | — |
| SimpleBench | — | 23% |
| Kagi LLM Benchmark | — | 45% |
| ARC-AGI-1 | 84% | — |
| CritPt | 8.3% | — |
| Chess Puzzles | 18% | — |
| Mystery Game Puzzles | 6% | — |
| DTBench | — | 61.4% |
| BIG-Bench Hard | — | 82.9% |
| ForecastBench | — | 59.9 |
| HellaSwag | — | 89.2% |
| PIQA | — | 85.9% |
| WinoGrande | — | 89.2% |
Math Inkling-Small leads
Inkling-Small: 45.1 (#77), Llama 3.1-405B: 18.4 (#290)
| Benchmark | Inkling-Small | Llama 3.1-405B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 90% | 9.7% |
| LMArena Math | 1459 | 1281 |
| FrontierMath (Tiers 1-3) | 46.3% | — |
| FrontierMath Tier 4 | 17.1% | — |
| ProofBench | 6% | — |
| Omni-MATH | — | 24.9% |
| MATH Level 5 | — | 49.8% |
Knowledge Inkling-Small leads
Inkling-Small: 48.2 (#77), Llama 3.1-405B: 30.4 (#227)
| Benchmark | Inkling-Small | Llama 3.1-405B |
|---|---|---|
| GPQA Diamond | 88.5% | 50.9% |
| LMArena Expert | 1442 | 1243 |
| SimpleQA Verified | 19.1% | — |
| MMLU-Pro | — | 72.3% |
| Confabulations | — | 17.6% |
| GPQA (HELM) | — | 52.2% |
| ARC (AI2) Challenge | — | 95.3% |
| MMLU | — | 84.5% |
| TriviaQA | — | 82.7% |
Multimodal Not comparable
Inkling-Small: 39.1 (#62), Llama 3.1-405B: —
| Benchmark | Inkling-Small | Llama 3.1-405B |
|---|---|---|
| LMArena Vision | 1235 | — |
Multilingual Inkling-Small leads
Inkling-Small: 51.7 (#104), Llama 3.1-405B: 40.7 (#214)
| Benchmark | Inkling-Small | Llama 3.1-405B |
|---|---|---|
| LMArena Non-English | 1402 | 1248 |
| LMArena Chinese | 1465 | 1242 |
| LMArena French | 1436 | 1279 |
| LMArena German | 1405 | 1252 |
| LMArena Japanese | 1405 | 1208 |
| LMArena Korean | 1363 | 1184 |
| LMArena Russian | 1391 | 1265 |
| LMArena Spanish | 1428 | 1260 |
Instruction Following Inkling-Small leads
Inkling-Small: 73.8 (#114), Llama 3.1-405B: 65.9 (#214)
| Benchmark | Inkling-Small | Llama 3.1-405B |
|---|---|---|
| LMArena Instruction Following | 1399 | 1259 |
| IFEval | — | 81.1% |
Long Context Inkling-Small leads
Inkling-Small: 42.7 (#118), Llama 3.1-405B: 38.4 (#197)
| Benchmark | Inkling-Small | Llama 3.1-405B |
|---|---|---|
| LMArena Longer Query | 1401 | 1266 |
Writing & Preference Inkling-Small leads
Inkling-Small: 59.6 (#107), Llama 3.1-405B: 38.9 (#251)
| Benchmark | Inkling-Small | Llama 3.1-405B |
|---|---|---|
| LMArena Text | 1414 | 1284 |
| LMArena Creative Writing | 1331 | 1262 |
| EQ-Bench Creative Writing | 1491 | 870 |
| LMArena Multi-Turn | 1418 | 1297 |
| WildBench | — | 78.3% |
Frequently asked questions
Is Inkling-Small better than Llama 3.1-405B?
Inkling-Small is the stronger model overall, scoring 46.5 to 30.7 on the Noometry Index.
Is Inkling-Small or Llama 3.1-405B better for coding?
Inkling-Small scores higher on coding benchmarks: 43.6 versus 33.1 in the Noometry coding category.
How many benchmarks do Inkling-Small and Llama 3.1-405B share?
21 benchmarks have published results for both models. Inkling-Small has 33 scored results on Noometry and Llama 3.1-405B has 42.