Model comparison
Inkling-Small vs Llama 3.2 3B
Inkling-Small is the stronger model overall, scoring 46.5 to 28.4 on the Noometry Index. Llama 3.2 3B costs 5.3× less per token, which makes it the better buy when Inkling-Small's lead doesn't matter for your workload.
Last verified . 14 shared benchmarks.
MetaSummary
- They share 14 benchmarks with published results for both. Inkling-Small scores higher in 8 categories and Llama 3.2 3B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Inkling-Small leads 59.8 to 24.9.
- Llama 3.2 3B is cheaper at $0.05 / $0.33 per million input/output tokens, against $0.45 / $1.20 for Inkling-Small.
- Inkling-Small accepts more context: 524K tokens versus 131K.
Side by side
| Inkling-Small | Llama 3.2 3B | |
|---|---|---|
| Provider | Thinking Machines Lab | Meta |
| Noometry Index | 46.5 | 28.4 |
| Released | 2026-07-15 | 2024-09-24 |
| Weights | Open | Open |
| Context window | 524K | 131K |
| Max output | 1.05M | 118K |
| Input $ / M tokens | $0.45 | $0.05 |
| Output $ / M tokens | $1.20 | $0.33 |
| Results tracked | 33 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Inkling-Small leads
Inkling-Small: 43.3 (#80), Llama 3.2 3B: 27.1 (#314)
| Benchmark | Inkling-Small | Llama 3.2 3B |
|---|---|---|
| LMArena Coding | 1451 | 1098 |
| LMArena WebDev | 1408 | — |
| SciCode | 48.7% | — |
| BigCodeBench Instruct | — | 23.4% |
| BigCodeBench Complete | — | 28.3% |
Agentic & Tool Use Not comparable
Inkling-Small: —, Llama 3.2 3B: 19.9 (#135)
| Benchmark | Inkling-Small | Llama 3.2 3B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 21.9% |
| BALROG | — | 10.1% |
Reasoning Inkling-Small leads
Inkling-Small: 39.0 (#53), Llama 3.2 3B: 20.1 (#225)
| Benchmark | Inkling-Small | Llama 3.2 3B |
|---|---|---|
| LMArena Hard Prompts | 1424 | 1095 |
| ARC-AGI-2 | 40.1% | — |
| ARC-AGI-1 | 84% | — |
| CritPt | 8.3% | — |
| Chess Puzzles | 18% | — |
| Mystery Game Puzzles | 6% | — |
| Epoch Capabilities Index | 150.13 | — |
Math Inkling-Small leads
Inkling-Small: 44.9 (#65), Llama 3.2 3B: 29.9 (#213)
| Benchmark | Inkling-Small | Llama 3.2 3B |
|---|---|---|
| LMArena Math | 1453 | 1126 |
| FrontierMath (Tiers 1-3) | 46.3% | — |
| FrontierMath Tier 4 | 17.1% | — |
| OTIS Mock AIME 2024-2025 | 90% | — |
| ProofBench | 6% | — |
Knowledge Inkling-Small leads
Inkling-Small: 48.3 (#71), Llama 3.2 3B: 29.1 (#232)
| Benchmark | Inkling-Small | Llama 3.2 3B |
|---|---|---|
| LMArena Expert | 1442 | 1090 |
| GPQA Diamond | 88.5% | — |
| SimpleQA Verified | 19.1% | — |
Multimodal Not comparable
Inkling-Small: 38.4 (#59), Llama 3.2 3B: —
| Benchmark | Inkling-Small | Llama 3.2 3B |
|---|---|---|
| LMArena Vision | 1235 | — |
Multilingual Inkling-Small leads
Inkling-Small: 51.7 (#102), Llama 3.2 3B: 26.4 (#276)
| Benchmark | Inkling-Small | Llama 3.2 3B |
|---|---|---|
| LMArena Non-English | 1400 | 1019 |
| LMArena Chinese | 1458 | 1017 |
| LMArena German | 1412 | 1056 |
| LMArena Russian | 1392 | 949 |
| LMArena French | 1438 | — |
| LMArena Japanese | 1399 | — |
| LMArena Korean | 1355 | — |
| LMArena Spanish | 1426 | — |
Instruction Following Inkling-Small leads
Inkling-Small: 73.9 (#112), Llama 3.2 3B: 56.3 (#270)
| Benchmark | Inkling-Small | Llama 3.2 3B |
|---|---|---|
| LMArena Instruction Following | 1397 | 1089 |
Long Context Inkling-Small leads
Inkling-Small: 42.7 (#113), Llama 3.2 3B: 33.4 (#256)
| Benchmark | Inkling-Small | Llama 3.2 3B |
|---|---|---|
| LMArena Longer Query | 1401 | 1101 |
Writing & Preference Inkling-Small leads
Inkling-Small: 59.8 (#102), Llama 3.2 3B: 24.9 (#302)
| Benchmark | Inkling-Small | Llama 3.2 3B |
|---|---|---|
| LMArena Text | 1413 | 1110 |
| LMArena Creative Writing | 1330 | 1094 |
| EQ-Bench Creative Writing | 1491 | 595 |
| LMArena Multi-Turn | 1420 | 1105 |
Frequently asked questions
Is Inkling-Small better than Llama 3.2 3B?
Inkling-Small is the stronger model overall, scoring 46.5 to 28.4 on the Noometry Index. Llama 3.2 3B costs 5.3× less per token, which makes it the better buy when Inkling-Small's lead doesn't matter for your workload.
Which is cheaper, Inkling-Small or Llama 3.2 3B?
Llama 3.2 3B is cheaper. It lists at $0.05 per million input tokens and $0.33 per million output tokens; Inkling-Small lists at $0.45 and $1.20.
Is Inkling-Small or Llama 3.2 3B better for coding?
Inkling-Small scores higher on coding benchmarks: 43.3 versus 27.1 in the Noometry coding category.
Which has the bigger context window?
Inkling-Small does, with 524K tokens against 131K.
How many benchmarks do Inkling-Small and Llama 3.2 3B share?
14 benchmarks have published results for both models. Inkling-Small has 33 scored results on Noometry and Llama 3.2 3B has 18.