Model comparison
Inkling vs Llama 3-8B
Inkling is the stronger model overall, scoring 44.1 to 25.5 on the Noometry Index.
Last verified . 22 shared benchmarks.
MetaSummary
- They share 22 benchmarks with published results for both. Inkling scores higher in 8 categories and Llama 3-8B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Inkling leads 55.1 to 7.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 88.9% for Inkling and 1.9% for Llama 3-8B.
Side by side
| Inkling | Llama 3-8B | |
|---|---|---|
| Provider | Thinking Machines Lab | Meta |
| Noometry Index | 44.1 | 25.5 |
| Released | 2026-07-15 | 2024-04-18 |
| Weights | Open | Open |
| Context window | 66K | — |
| Max output | 66K | — |
| Input $ / M tokens | $1.87 | — |
| Output $ / M tokens | $4.68 | — |
| Results tracked | 41 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Inkling leads
Inkling: 34.5 (#234), Llama 3-8B: 31.0 (#289)
| Benchmark | Inkling | Llama 3-8B |
|---|---|---|
| LMArena Coding | 1464 | 1152 |
| FrontierCode | 14% | — |
| LMArena WebDev | 1413 | — |
| FrontierSWE | 4.1% | — |
| SciCode | 47% | — |
| WeirdML | 32.3% | — |
| BigCodeBench Instruct | — | 31.9% |
| BigCodeBench Complete | — | 36.9% |
| ALE-Bench | 946 | — |
| HumanEval+ | — | 56.7% |
| MBPP+ | — | 54.8% |
Agentic & Tool Use Not comparable
Inkling: 29.6 (#85), Llama 3-8B: —
| Benchmark | Inkling | Llama 3-8B |
|---|---|---|
| APEX-Agents | 33.8% | — |
| τ²-bench Banking | 25% | — |
Reasoning Inkling leads
Inkling: 40.4 (#56), Llama 3-8B: 14.3 (#326)
| Benchmark | Inkling | Llama 3-8B |
|---|---|---|
| Chess Puzzles | 21% | 0% |
| LMArena Hard Prompts | 1451 | 1133 |
| DTBench | 87.5% | 43.9% |
| Epoch Capabilities Index | 148.54 | 116.45 |
| ARC-AGI-2 | 36.5% | — |
| SimpleBench | 50% | — |
| ARC-AGI-1 | 79.5% | — |
| CritPt | 5.4% | — |
| LMCA | 37.6% | — |
| Adversarial NLI | — | 57.3% |
| ForecastBench | — | 58.6 |
| WinoGrande | — | 75.7% |
Math Inkling leads
Inkling: 31.3 (#225), Llama 3-8B: 8.8 (#323)
| Benchmark | Inkling | Llama 3-8B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 1.9% |
| LMArena Math | 1479 | 1151 |
| FrontierMath (Tiers 1-3) | 33.3% | — |
| FrontierMath Tier 4 | 4.9% | — |
| ProofBench | 0% | — |
| MATH Level 5 | — | 6.1% |
Knowledge Inkling leads
Inkling: 55.1 (#49), Llama 3-8B: 7.8 (#308)
| Benchmark | Inkling | Llama 3-8B |
|---|---|---|
| GPQA Diamond | 88.3% | 26.1% |
| LMArena Expert | 1465 | 1113 |
| SimpleQA Verified | 40.3% | — |
| ARC (AI2) Challenge | — | 82.8% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 82.6% |
| TriviaQA | — | 67.7% |
Multilingual Inkling leads
Inkling: 54.0 (#52), Llama 3-8B: 30.8 (#261)
| Benchmark | Inkling | Llama 3-8B |
|---|---|---|
| LMArena Non-English | 1434 | 1098 |
| LMArena Chinese | 1490 | 1076 |
| LMArena French | 1458 | 1159 |
| LMArena German | 1446 | 1104 |
| LMArena Japanese | 1429 | 967 |
| LMArena Korean | 1404 | 1004 |
| LMArena Russian | 1429 | 1109 |
| LMArena Spanish | 1448 | 1173 |
Instruction Following Inkling leads
Inkling: 75.1 (#71), Llama 3-8B: 58.4 (#260)
| Benchmark | Inkling | Llama 3-8B |
|---|---|---|
| LMArena Instruction Following | 1426 | 1127 |
Long Context Inkling leads
Inkling: 43.8 (#86), Llama 3-8B: 34.2 (#251)
| Benchmark | Inkling | Llama 3-8B |
|---|---|---|
| LMArena Longer Query | 1434 | 1128 |
Writing & Preference Inkling leads
Inkling: 65.2 (#51), Llama 3-8B: 37.5 (#256)
| Benchmark | Inkling | Llama 3-8B |
|---|---|---|
| LMArena Text | 1441 | 1166 |
| LMArena Creative Writing | 1387 | 1150 |
| LMArena Multi-Turn | 1436 | 1152 |
| EQ-Bench Creative Writing | 1611 | — |
| EQ-Bench 4 | 1226 | — |
Frequently asked questions
Is Inkling better than Llama 3-8B?
Inkling is the stronger model overall, scoring 44.1 to 25.5 on the Noometry Index.
Is Inkling or Llama 3-8B better for coding?
Inkling scores higher on coding benchmarks: 34.5 versus 31.0 in the Noometry coding category.
How many benchmarks do Inkling and Llama 3-8B share?
22 benchmarks have published results for both models. Inkling has 41 scored results on Noometry and Llama 3-8B has 34.