Model comparison
Inkling vs Phi 3 Mini 4k Instruct
Inkling is the stronger model overall, scoring 44.1 to 27.9 on the Noometry Index.
Last verified . 18 shared benchmarks.
MicrosoftSummary
- They share 18 benchmarks with published results for both. Inkling scores higher in 8 categories and Phi 3 Mini 4k Instruct in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Inkling leads 65.2 to 27.6.
- The biggest single-benchmark swing is Chess Puzzles: 21% for Inkling and 0% for Phi 3 Mini 4k Instruct.
Side by side
| Inkling | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | Thinking Machines Lab | Microsoft |
| Noometry Index | 44.1 | 27.9 |
| Released | 2026-07-15 | 2024-04-23 |
| Weights | Open | Open |
| Context window | 66K | — |
| Max output | 66K | — |
| Input $ / M tokens | $1.87 | — |
| Output $ / M tokens | $4.68 | — |
| Results tracked | 41 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Inkling leads
Inkling: 34.5 (#234), Phi 3 Mini 4k Instruct: 26.6 (#323)
| Benchmark | Inkling | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Coding | 1464 | 1093 |
| FrontierCode | 14% | — |
| LMArena WebDev | 1413 | — |
| FrontierSWE | 4.1% | — |
| SciCode | 47% | — |
| WeirdML | 32.3% | — |
| LiveBench Coding | — | 15.5% |
| ALE-Bench | 946 | — |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 54.2% |
Agentic & Tool Use Not comparable
Inkling: 29.6 (#85), Phi 3 Mini 4k Instruct: —
| Benchmark | Inkling | Phi 3 Mini 4k Instruct |
|---|---|---|
| APEX-Agents | 33.8% | — |
| τ²-bench Banking | 25% | — |
Reasoning Inkling leads
Inkling: 40.4 (#56), Phi 3 Mini 4k Instruct: 14.1 (#328)
| Benchmark | Inkling | Phi 3 Mini 4k Instruct |
|---|---|---|
| Chess Puzzles | 21% | 0% |
| LMArena Hard Prompts | 1451 | 1072 |
| ARC-AGI-2 | 36.5% | — |
| SimpleBench | 50% | — |
| ARC-AGI-1 | 79.5% | — |
| CritPt | 5.4% | — |
| LiveBench Reasoning | — | 26.8% |
| DTBench | 87.5% | — |
| LiveBench Data Analysis | — | 34.7% |
| LMCA | 37.6% | — |
| Adversarial NLI | — | 52.8% |
| BIG-Bench Hard | — | 71.7% |
| Epoch Capabilities Index | 148.54 | — |
| HellaSwag | — | 76.7% |
| LiveBench | — | 22.4% |
| WinoGrande | — | 70.8% |
Math Inkling leads
Inkling: 31.3 (#225), Phi 3 Mini 4k Instruct: 26.6 (#257)
| Benchmark | Inkling | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Math | 1479 | 1111 |
| FrontierMath (Tiers 1-3) | 33.3% | — |
| FrontierMath Tier 4 | 4.9% | — |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| ProofBench | 0% | — |
| LiveBench Math | — | 15.7% |
Knowledge Inkling leads
Inkling: 55.1 (#49), Phi 3 Mini 4k Instruct: 28.5 (#246)
| Benchmark | Inkling | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1465 | 1045 |
| GPQA Diamond | 88.3% | — |
| SimpleQA Verified | 40.3% | — |
| ARC (AI2) Challenge | — | 84.9% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 64% |
Multilingual Inkling leads
Inkling: 54.0 (#52), Phi 3 Mini 4k Instruct: 26.3 (#280)
| Benchmark | Inkling | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1434 | 1021 |
| LMArena Chinese | 1490 | 1021 |
| LMArena French | 1458 | 1076 |
| LMArena German | 1446 | 1044 |
| LMArena Japanese | 1429 | 935 |
| LMArena Korean | 1404 | 905 |
| LMArena Russian | 1429 | 1022 |
| LMArena Spanish | 1448 | 1085 |
Instruction Following Inkling leads
Inkling: 75.1 (#71), Phi 3 Mini 4k Instruct: 47.7 (#303)
| Benchmark | Inkling | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1426 | 1053 |
| LiveBench Instruction Following | — | 39.1% |
Long Context Inkling leads
Inkling: 43.8 (#86), Phi 3 Mini 4k Instruct: 31.7 (#276)
| Benchmark | Inkling | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1434 | 1044 |
Writing & Preference Inkling leads
Inkling: 65.2 (#51), Phi 3 Mini 4k Instruct: 27.6 (#300)
| Benchmark | Inkling | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1441 | 1073 |
| LMArena Creative Writing | 1387 | 1037 |
| LMArena Multi-Turn | 1436 | 1018 |
| EQ-Bench Creative Writing | 1611 | — |
| EQ-Bench 4 | 1226 | — |
| LiveBench Language | — | 9.2% |
Frequently asked questions
Is Inkling better than Phi 3 Mini 4k Instruct?
Inkling is the stronger model overall, scoring 44.1 to 27.9 on the Noometry Index.
Is Inkling or Phi 3 Mini 4k Instruct better for coding?
Inkling scores higher on coding benchmarks: 34.5 versus 26.6 in the Noometry coding category.
How many benchmarks do Inkling and Phi 3 Mini 4k Instruct share?
18 benchmarks have published results for both models. Inkling has 41 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.