Model comparison
Inkling vs Step 3
Inkling is the stronger model overall, scoring 44.1 to 39.9 on the Noometry Index.
Last verified . 15 shared benchmarks.
StepFunSummary
- They share 15 benchmarks with published results for both. Inkling scores higher in 6 categories and Step 3 in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Inkling leads 55.2 to 36.2.
Side by side
| Inkling | Step 3 | |
|---|---|---|
| Provider | Thinking Machines Lab | StepFun |
| Noometry Index | 44.1 | 39.9 |
| Released | 2026-07-15 | — |
| Weights | Open | Open |
| Context window | 66K | — |
| Max output | 66K | — |
| Input $ / M tokens | $1.87 | — |
| Output $ / M tokens | $4.68 | — |
| Results tracked | 41 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3 leads
Inkling: 34.1 (#229), Step 3: 39.5 (#144)
| Benchmark | Inkling | Step 3 |
|---|---|---|
| LMArena Coding | 1463 | 1367 |
| FrontierCode | 14% | — |
| LMArena WebDev | 1412 | — |
| FrontierSWE | 4.1% | — |
| SciCode | 47% | — |
| WeirdML | 32.3% | — |
| ALE-Bench | 946 | — |
Agentic & Tool Use Not comparable
Inkling: 29.5 (#76), Step 3: —
| Benchmark | Inkling | Step 3 |
|---|---|---|
| APEX-Agents | 33.8% | — |
| τ²-bench Banking | 25% | — |
Reasoning Inkling leads
Inkling: 40.6 (#48), Step 3: 27.6 (#102)
| Benchmark | Inkling | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1452 | 1355 |
| ARC-AGI-2 | 36.5% | — |
| SimpleBench | 50% | — |
| Kagi LLM Benchmark | — | 62.3% |
| ARC-AGI-1 | 79.5% | — |
| CritPt | 5.4% | — |
| Chess Puzzles | 21% | — |
| DTBench | 87.5% | — |
| LMCA | 37.6% | — |
| Epoch Capabilities Index | 148.6 | — |
Math Step 3 leads
Inkling: 31.2 (#198), Step 3: 34.9 (#148)
| Benchmark | Inkling | Step 3 |
|---|---|---|
| LMArena Math | 1482 | 1366 |
| FrontierMath (Tiers 1-3) | 33.3% | — |
| FrontierMath Tier 4 | 4.9% | — |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| ProofBench | 0% | — |
Knowledge Inkling leads
Inkling: 55.2 (#44), Step 3: 36.2 (#161)
| Benchmark | Inkling | Step 3 |
|---|---|---|
| LMArena Expert | 1468 | 1333 |
| GPQA Diamond | 88.3% | — |
| SimpleQA Verified | 40.3% | — |
Multimodal Not comparable
Inkling: —, Step 3: 34.9 (#81)
| Benchmark | Inkling | Step 3 |
|---|---|---|
| LMArena Vision | — | 1176 |
Multilingual Inkling leads
Inkling: 54.2 (#49), Step 3: 46.5 (#154)
| Benchmark | Inkling | Step 3 |
|---|---|---|
| LMArena Non-English | 1434 | 1327 |
| LMArena Chinese | 1495 | 1396 |
| LMArena German | 1456 | 1371 |
| LMArena Korean | 1401 | 1271 |
| LMArena Russian | 1430 | 1331 |
| LMArena Spanish | 1442 | 1372 |
| LMArena French | 1453 | — |
| LMArena Japanese | 1428 | — |
Instruction Following Inkling leads
Inkling: 75.3 (#65), Step 3: 70.6 (#159)
| Benchmark | Inkling | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1427 | 1332 |
Long Context Inkling leads
Inkling: 43.9 (#79), Step 3: 40.3 (#152)
| Benchmark | Inkling | Step 3 |
|---|---|---|
| LMArena Longer Query | 1436 | 1326 |
Writing & Preference Inkling leads
Inkling: 65.5 (#48), Step 3: 54.5 (#146)
| Benchmark | Inkling | Step 3 |
|---|---|---|
| LMArena Text | 1443 | 1351 |
| LMArena Creative Writing | 1390 | 1322 |
| LMArena Multi-Turn | 1436 | 1340 |
| EQ-Bench Creative Writing | 1611 | — |
| EQ-Bench 4 | 1226 | — |
Frequently asked questions
Is Inkling better than Step 3?
Inkling is the stronger model overall, scoring 44.1 to 39.9 on the Noometry Index.
Is Inkling or Step 3 better for coding?
Step 3 scores higher on coding benchmarks: 39.5 versus 34.1 in the Noometry coding category.
How many benchmarks do Inkling and Step 3 share?
15 benchmarks have published results for both models. Inkling has 41 scored results on Noometry and Step 3 has 17.