Model comparison
GPT-4.1 mini vs Inkling
Inkling is the stronger model overall, scoring 44.1 to 33.5 on the Noometry Index. GPT-4.1 mini costs 3.7× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.
Last verified . 31 shared benchmarks.
Summary
- They share 31 benchmarks with published results for both. GPT-4.1 mini scores higher in 1 category and Inkling in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling leads 40.6 to 10.9.
- The biggest single-benchmark swing is ARC-AGI-1: 3.5% for GPT-4.1 mini and 79.5% for Inkling.
- GPT-4.1 mini is cheaper at $0.40 / $1.60 per million input/output tokens, against $1.87 / $4.68 for Inkling.
- GPT-4.1 mini accepts more context: 1.05M tokens versus 66K.
- Inkling has downloadable open weights; the other is API-only.
Side by side
| GPT-4.1 mini | Inkling | |
|---|---|---|
| Provider | OpenAI | Thinking Machines Lab |
| Noometry Index | 33.5 | 44.1 |
| Released | 2025-04-14 | 2026-07-15 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 66K |
| Max output | 33K | 66K |
| Input $ / M tokens | $0.40 | $1.87 |
| Output $ / M tokens | $1.60 | $4.68 |
| Results tracked | 47 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Inkling leads
GPT-4.1 mini: 30.4 (#286), Inkling: 34.1 (#229)
| Benchmark | GPT-4.1 mini | Inkling |
|---|---|---|
| SciCode | 40.4% | 47% |
| WeirdML | 37.6% | 32.3% |
| LMArena Coding | 1367 | 1463 |
| FrontierCode | — | 14% |
| SWE-bench Verified (bash only) | 23.9% | — |
| Aider Polyglot | 32.4% | — |
| LMArena WebDev | — | 1412 |
| FrontierSWE | — | 4.1% |
| BigCodeBench Instruct | 48.9% | — |
| CadEval | 16% | — |
| ALE-Bench | — | 946 |
Agentic & Tool Use GPT-4.1 mini leads
GPT-4.1 mini: 33.0 (#52), Inkling: 29.5 (#76)
| Benchmark | GPT-4.1 mini | Inkling |
|---|---|---|
| APEX-Agents | — | 33.8% |
| Berkeley Function Calling Leaderboard | 50.5% | — |
| τ²-bench Banking | — | 25% |
Reasoning Inkling leads
GPT-4.1 mini: 10.9 (#335), Inkling: 40.6 (#48)
| Benchmark | GPT-4.1 mini | Inkling |
|---|---|---|
| ARC-AGI-2 | 0% | 36.5% |
| ARC-AGI-1 | 3.5% | 79.5% |
| CritPt | 0% | 5.4% |
| Chess Puzzles | 7% | 21% |
| LMArena Hard Prompts | 1349 | 1452 |
| DTBench | 68.8% | 87.5% |
| LMCA | 21.1% | 37.6% |
| Epoch Capabilities Index | 135.01 | 148.6 |
| SimpleBench | — | 50% |
| Kagi LLM Benchmark | 48.6% | — |
| Mystery Game Puzzles | 7% | — |
Math Inkling leads
GPT-4.1 mini: 23.1 (#264), Inkling: 31.2 (#198)
| Benchmark | GPT-4.1 mini | Inkling |
|---|---|---|
| FrontierMath (Tiers 1-3) | 6.7% | 33.3% |
| OTIS Mock AIME 2024-2025 | 44.7% | 88.9% |
| LMArena Math | 1343 | 1482 |
| FrontierMath Tier 4 | — | 4.9% |
| ProofBench | — | 0% |
| Omni-MATH | 49.1% | — |
| MATH Level 5 | 87.3% | — |
| FrontierMath (Feb 2025 set) | 4.5% | — |
Knowledge Inkling leads
GPT-4.1 mini: 34.7 (#183), Inkling: 55.2 (#44)
| Benchmark | GPT-4.1 mini | Inkling |
|---|---|---|
| GPQA Diamond | 65.8% | 88.3% |
| SimpleQA Verified | 12.7% | 40.3% |
| LMArena Expert | 1338 | 1468 |
| MMLU-Pro | 78.3% | — |
| GPQA (HELM) | 61.4% | — |
Multimodal Not comparable
GPT-4.1 mini: 35.2 (#77), Inkling: —
| Benchmark | GPT-4.1 mini | Inkling |
|---|---|---|
| LMArena Vision | 1181 | — |
Multilingual Inkling leads
GPT-4.1 mini: 45.9 (#161), Inkling: 54.2 (#49)
| Benchmark | GPT-4.1 mini | Inkling |
|---|---|---|
| LMArena Non-English | 1318 | 1434 |
| LMArena Chinese | 1329 | 1495 |
| LMArena French | 1357 | 1453 |
| LMArena German | 1349 | 1456 |
| LMArena Japanese | 1291 | 1428 |
| LMArena Korean | 1298 | 1401 |
| LMArena Russian | 1324 | 1430 |
| LMArena Spanish | 1319 | 1442 |
Instruction Following Inkling leads
GPT-4.1 mini: 73.9 (#113), Inkling: 75.3 (#65)
| Benchmark | GPT-4.1 mini | Inkling |
|---|---|---|
| LMArena Instruction Following | 1332 | 1427 |
| IFEval | 90.4% | — |
Long Context Inkling leads
GPT-4.1 mini: 31.8 (#270), Inkling: 43.9 (#79)
| Benchmark | GPT-4.1 mini | Inkling |
|---|---|---|
| LMArena Longer Query | 1344 | 1436 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference Inkling leads
GPT-4.1 mini: 48.8 (#193), Inkling: 65.5 (#48)
| Benchmark | GPT-4.1 mini | Inkling |
|---|---|---|
| LMArena Text | 1341 | 1443 |
| LMArena Creative Writing | 1300 | 1390 |
| EQ-Bench Creative Writing | 1147 | 1611 |
| LMArena Multi-Turn | 1354 | 1436 |
| WildBench | 83.8% | — |
| EQ-Bench 4 | — | 1226 |
Frequently asked questions
Is GPT-4.1 mini better than Inkling?
Inkling is the stronger model overall, scoring 44.1 to 33.5 on the Noometry Index. GPT-4.1 mini costs 3.7× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.
Which is cheaper, GPT-4.1 mini or Inkling?
GPT-4.1 mini is cheaper. It lists at $0.40 per million input tokens and $1.60 per million output tokens; Inkling lists at $1.87 and $4.68.
Is GPT-4.1 mini or Inkling better for coding?
Inkling scores higher on coding benchmarks: 34.1 versus 30.4 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 mini does, with 1.05M tokens against 66K.
How many benchmarks do GPT-4.1 mini and Inkling share?
31 benchmarks have published results for both models. GPT-4.1 mini has 47 scored results on Noometry and Inkling has 41.