Model comparison
GPT-4o vs Inkling
Inkling is the stronger model overall, scoring 44.1 to 28.6 on the Noometry Index.
Last verified . 30 shared benchmarks.
Summary
- They share 30 benchmarks with published results for both. GPT-4o scores higher in 0 categories and Inkling in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling leads 40.4 to 9.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.4% for GPT-4o and 88.9% for Inkling.
- Inkling is cheaper at $1.87 / $4.68 per million input/output tokens, against $2.50 / $10 for GPT-4o.
- GPT-4o accepts more context: 128K tokens versus 66K.
- Inkling has downloadable open weights; the other is API-only.
Side by side
| GPT-4o | Inkling | |
|---|---|---|
| Provider | OpenAI | Thinking Machines Lab |
| Noometry Index | 28.6 | 44.1 |
| Released | 2024-05-13 | 2026-07-15 |
| Weights | Proprietary | Open |
| Context window | 128K | 66K |
| Max output | 16K | 66K |
| Input $ / M tokens | $2.50 | $1.87 |
| Output $ / M tokens | $10 | $4.68 |
| Results tracked | 72 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Inkling leads
GPT-4o: 24.8 (#328), Inkling: 34.5 (#234)
| Benchmark | GPT-4o | Inkling |
|---|---|---|
| WeirdML | 25.1% | 32.3% |
| LMArena Coding | 1297 | 1464 |
| SWE-bench Verified | 31% | — |
| FrontierCode | — | 14% |
| SWE-bench Verified (bash only) | 21.6% | — |
| Aider Polyglot | 45.3% | — |
| LMArena WebDev | — | 1413 |
| FrontierSWE | — | 4.1% |
| SciCode | — | 47% |
| GSO | 0% | — |
| BigCodeBench Instruct | 51.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 61.1% | — |
| CadEval | 26% | — |
| ALE-Bench | — | 946 |
| HumanEval+ | 87.2% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use Inkling leads
GPT-4o: 21.0 (#141), Inkling: 29.6 (#85)
| Benchmark | GPT-4o | Inkling |
|---|---|---|
| APEX-Agents | — | 33.8% |
| GDPval | 9.9% | — |
| TheAgentCompany | 8.6% | — |
| τ²-bench Banking | — | 25% |
| Cybench | 12.5% | — |
| BALROG | 32.3% | — |
| LMArena Search | 1006 | — |
| METR Time Horizons | 40.8% | — |
Reasoning Inkling leads
GPT-4o: 9.4 (#343), Inkling: 40.4 (#56)
| Benchmark | GPT-4o | Inkling |
|---|---|---|
| ARC-AGI-2 | 0% | 36.5% |
| SimpleBench | 17.8% | 50% |
| ARC-AGI-1 | 4.5% | 79.5% |
| CritPt | 0% | 5.4% |
| Chess Puzzles | 13% | 21% |
| LMArena Hard Prompts | 1281 | 1451 |
| DTBench | 64.5% | 87.5% |
| LMCA | 16.6% | 37.6% |
| Epoch Capabilities Index | 128.97 | 148.54 |
| EnigmaEval | 0.8% | — |
| LiveBench Reasoning | 55.8% | — |
| LiveBench Data Analysis | 60.9% | — |
| ForecastBench | 57.7 | — |
| LiveBench | 55.3% | — |
Math Inkling leads
GPT-4o: 10.6 (#312), Inkling: 31.3 (#225)
| Benchmark | GPT-4o | Inkling |
|---|---|---|
| FrontierMath (Tiers 1-3) | 0.4% | 33.3% |
| OTIS Mock AIME 2024-2025 | 6.4% | 88.9% |
| LMArena Math | 1285 | 1479 |
| FrontierMath Tier 4 | — | 4.9% |
| ProofBench | — | 0% |
| Omni-MATH | 29.3% | — |
| LiveBench Math | 49.5% | — |
| MATH Level 5 | 53.3% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Inkling leads
GPT-4o: 28.8 (#242), Inkling: 55.1 (#49)
| Benchmark | GPT-4o | Inkling |
|---|---|---|
| GPQA Diamond | 49.2% | 88.3% |
| SimpleQA Verified | 26% | 40.3% |
| LMArena Expert | 1250 | 1465 |
| Humanity's Last Exam | 2.7% | — |
| MMLU-Pro | 71.3% | — |
| Confabulations | 15.3% | — |
| Vectara Hallucination Rate | 9.6% | — |
| GPQA (HELM) | 52% | — |
| MMLU | 88.1% | — |
Multimodal Not comparable
GPT-4o: 34.5 (#91), Inkling: —
| Benchmark | GPT-4o | Inkling |
|---|---|---|
| LMArena Vision | 1137 | — |
| Video-MME | 71.9% | — |
| GeoBench | 71% | — |
| VPCT | 40% | — |
| ScienceQA | 88.5% | — |
Multilingual Inkling leads
GPT-4o: 43.2 (#186), Inkling: 54.0 (#52)
| Benchmark | GPT-4o | Inkling |
|---|---|---|
| LMArena Non-English | 1283 | 1434 |
| LMArena Chinese | 1277 | 1490 |
| LMArena French | 1304 | 1458 |
| LMArena German | 1282 | 1446 |
| LMArena Japanese | 1257 | 1429 |
| LMArena Korean | 1234 | 1404 |
| LMArena Russian | 1286 | 1429 |
| LMArena Spanish | 1292 | 1448 |
Instruction Following Inkling leads
GPT-4o: 66.6 (#207), Inkling: 75.1 (#71)
| Benchmark | GPT-4o | Inkling |
|---|---|---|
| LMArena Instruction Following | 1278 | 1426 |
| LiveBench Instruction Following | 68.6% | — |
| IFEval | 81.7% | — |
Long Context Inkling leads
GPT-4o: 39.4 (#179), Inkling: 43.8 (#86)
| Benchmark | GPT-4o | Inkling |
|---|---|---|
| LMArena Longer Query | 1289 | 1434 |
| Fiction.LiveBench | 66.7% | — |
Writing & Preference Inkling leads
GPT-4o: 52.6 (#166), Inkling: 65.2 (#51)
| Benchmark | GPT-4o | Inkling |
|---|---|---|
| LMArena Text | 1300 | 1441 |
| LMArena Creative Writing | 1292 | 1387 |
| LMArena Multi-Turn | 1302 | 1436 |
| Short-Story Creative Writing | 81.8% | — |
| EQ-Bench Creative Writing | — | 1611 |
| WildBench | 82.8% | — |
| EQ-Bench 4 | — | 1226 |
| LiveBench Language | 47.6% | — |
Frequently asked questions
Is GPT-4o better than Inkling?
Inkling is the stronger model overall, scoring 44.1 to 28.6 on the Noometry Index.
Which is cheaper, GPT-4o or Inkling?
Inkling is cheaper. It lists at $1.87 per million input tokens and $4.68 per million output tokens; GPT-4o lists at $2.50 and $10.
Is GPT-4o or Inkling better for coding?
Inkling scores higher on coding benchmarks: 34.5 versus 24.8 in the Noometry coding category.
Which has the bigger context window?
GPT-4o does, with 128K tokens against 66K.
How many benchmarks do GPT-4o and Inkling share?
30 benchmarks have published results for both models. GPT-4o has 72 scored results on Noometry and Inkling has 41.