Model comparison
Inkling-Small vs Phi 3 Medium 4k Instruct
Inkling-Small is the stronger model overall, scoring 46.5 to 33.0 on the Noometry Index.
Last verified . 17 shared benchmarks.
MicrosoftSummary
- They share 17 benchmarks with published results for both. Inkling-Small scores higher in 8 categories and Phi 3 Medium 4k Instruct in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Inkling-Small leads 59.6 to 34.1.
Side by side
| Inkling-Small | Phi 3 Medium 4k Instruct | |
|---|---|---|
| Provider | Thinking Machines Lab | Microsoft |
| Noometry Index | 46.5 | 33.0 |
| Released | 2026-07-15 | — |
| Weights | Open | Open |
| Context window | 524K | — |
| Max output | 1.05M | — |
| Input $ / M tokens | $0.45 | — |
| Output $ / M tokens | $1.20 | — |
| Results tracked | 33 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Inkling-Small leads
Inkling-Small: 43.6 (#85), Phi 3 Medium 4k Instruct: 32.9 (#268)
| Benchmark | Inkling-Small | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Coding | 1451 | 1130 |
| LMArena WebDev | 1409 | — |
| SciCode | 48.7% | — |
Reasoning Inkling-Small leads
Inkling-Small: 38.6 (#63), Phi 3 Medium 4k Instruct: 21.7 (#214)
| Benchmark | Inkling-Small | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1423 | 1127 |
| ARC-AGI-2 | 40.1% | — |
| ARC-AGI-1 | 84% | — |
| CritPt | 8.3% | — |
| Chess Puzzles | 18% | — |
| Mystery Game Puzzles | 6% | — |
| Epoch Capabilities Index | 150.15 | — |
Math Inkling-Small leads
Inkling-Small: 45.1 (#77), Phi 3 Medium 4k Instruct: 33.4 (#202)
| Benchmark | Inkling-Small | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Math | 1459 | 1173 |
| 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: 48.2 (#77), Phi 3 Medium 4k Instruct: 30.2 (#229)
| Benchmark | Inkling-Small | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Expert | 1442 | 1108 |
| GPQA Diamond | 88.5% | — |
| SimpleQA Verified | 19.1% | — |
Multimodal Not comparable
Inkling-Small: 39.1 (#62), Phi 3 Medium 4k Instruct: —
| Benchmark | Inkling-Small | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Vision | 1235 | — |
Multilingual Inkling-Small leads
Inkling-Small: 51.7 (#104), Phi 3 Medium 4k Instruct: 30.6 (#263)
| Benchmark | Inkling-Small | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Non-English | 1402 | 1094 |
| LMArena Chinese | 1465 | 1108 |
| LMArena French | 1436 | 1110 |
| LMArena German | 1405 | 1101 |
| LMArena Japanese | 1405 | 1042 |
| LMArena Korean | 1363 | 954 |
| LMArena Russian | 1391 | 1145 |
| LMArena Spanish | 1428 | 1095 |
Instruction Following Inkling-Small leads
Inkling-Small: 73.8 (#114), Phi 3 Medium 4k Instruct: 57.6 (#268)
| Benchmark | Inkling-Small | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1399 | 1113 |
Long Context Inkling-Small leads
Inkling-Small: 42.7 (#118), Phi 3 Medium 4k Instruct: 34.0 (#255)
| Benchmark | Inkling-Small | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1401 | 1120 |
Writing & Preference Inkling-Small leads
Inkling-Small: 59.6 (#107), Phi 3 Medium 4k Instruct: 34.1 (#272)
| Benchmark | Inkling-Small | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Text | 1414 | 1138 |
| LMArena Creative Writing | 1331 | 1107 |
| LMArena Multi-Turn | 1418 | 1088 |
| EQ-Bench Creative Writing | 1491 | — |
Frequently asked questions
Is Inkling-Small better than Phi 3 Medium 4k Instruct?
Inkling-Small is the stronger model overall, scoring 46.5 to 33.0 on the Noometry Index.
Is Inkling-Small or Phi 3 Medium 4k Instruct better for coding?
Inkling-Small scores higher on coding benchmarks: 43.6 versus 32.9 in the Noometry coding category.
How many benchmarks do Inkling-Small and Phi 3 Medium 4k Instruct share?
17 benchmarks have published results for both models. Inkling-Small has 33 scored results on Noometry and Phi 3 Medium 4k Instruct has 17.