Model comparison
GPT-5 vs Inkling
GPT-5 is the stronger model overall, scoring 50.9 to 44.1 on the Noometry Index.
Last verified . 36 shared benchmarks.
Summary
- They share 36 benchmarks with published results for both. GPT-5 scores higher in 5 categories and Inkling in 4 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in long context, where GPT-5 leads 69.5 to 43.8.
- The biggest single-benchmark swing is WeirdML: 60.7% for GPT-5 and 32.3% for Inkling.
- Inkling is cheaper at $1.87 / $4.68 per million input/output tokens, against $1.25 / $10 for GPT-5.
- GPT-5 accepts more context: 400K tokens versus 66K.
- Inkling has downloadable open weights; the other is API-only.
Side by side
| GPT-5 | Inkling | |
|---|---|---|
| Provider | OpenAI | Thinking Machines Lab |
| Noometry Index | 50.9 | 44.1 |
| Released | 2025-08-07 | 2026-07-15 |
| Weights | Proprietary | Open |
| Context window | 400K | 66K |
| Max output | 128K | 66K |
| Input $ / M tokens | $1.25 | $1.87 |
| Output $ / M tokens | $10 | $4.68 |
| Results tracked | 69 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 leads
GPT-5: 50.3 (#47), Inkling: 34.5 (#234)
| Benchmark | GPT-5 | Inkling |
|---|---|---|
| LMArena WebDev | 1418 | 1413 |
| SciCode | 42.9% | 47% |
| WeirdML | 60.7% | 32.3% |
| LMArena Coding | 1436 | 1464 |
| ALE-Bench | 1,162 | 946 |
| SWE-bench Verified | 73.6% | — |
| FrontierCode | — | 14% |
| SWE-bench Verified (bash only) | 65% | — |
| Aider Polyglot | 88% | — |
| FrontierSWE | — | 4.1% |
| GSO | 6.9% | — |
| AlgoTune | 1.67 | — |
Agentic & Tool Use GPT-5 leads
GPT-5: 33.1 (#56), Inkling: 29.6 (#85)
| Benchmark | GPT-5 | Inkling |
|---|---|---|
| Terminal-Bench | 49.6% | — |
| APEX-Agents | — | 33.8% |
| GDPval | 34.8% | — |
| Remote Labor Index | 1.7% | — |
| τ²-bench Banking | — | 25% |
| DeepResearch Bench | 49.6% | — |
| BALROG | 32.8% | — |
| LMArena Search | 1133 | — |
| METR Time Horizons | 69.6% | — |
Reasoning Inkling leads
GPT-5: 38.3 (#64), Inkling: 40.4 (#56)
| Benchmark | GPT-5 | Inkling |
|---|---|---|
| ARC-AGI-2 | 9.9% | 36.5% |
| SimpleBench | 56.7% | 50% |
| ARC-AGI-1 | 65.7% | 79.5% |
| CritPt | 12.6% | 5.4% |
| Chess Puzzles | 37% | 21% |
| LMArena Hard Prompts | 1416 | 1451 |
| DTBench | 90.7% | 87.5% |
| LMCA | 40% | 37.6% |
| Epoch Capabilities Index | 150 | 148.54 |
| Kagi LLM Benchmark | 72.7% | — |
| EnigmaEval | 10.5% | — |
| EBR-Bench | 12.7% | — |
| Mystery Game Puzzles | 23% | — |
| ForecastBench | 61.4 | — |
Math GPT-5 leads
GPT-5: 55.0 (#44), Inkling: 31.3 (#225)
| Benchmark | GPT-5 | Inkling |
|---|---|---|
| FrontierMath (Tiers 1-3) | 55.4% | 33.3% |
| FrontierMath Tier 4 | 22% | 4.9% |
| OTIS Mock AIME 2024-2025 | 91.4% | 88.9% |
| ProofBench | 18% | 0% |
| LMArena Math | 1407 | 1479 |
| Omni-MATH | 64.7% | — |
| MATH Level 5 | 98.1% | — |
| FrontierMath (Feb 2025 set) | 32.4% | — |
| FrontierMath Tier 4 (v1) | 12.5% | — |
Knowledge GPT-5 leads
GPT-5: 56.6 (#43), Inkling: 55.1 (#49)
| Benchmark | GPT-5 | Inkling |
|---|---|---|
| GPQA Diamond | 86.2% | 88.3% |
| SimpleQA Verified | 50.1% | 40.3% |
| LMArena Expert | 1419 | 1465 |
| Humanity's Last Exam | 25.3% | — |
| MMLU-Pro | 86.3% | — |
| Confabulations | 10.3% | — |
| Vectara Hallucination Rate | 14.7% | — |
| GPQA (HELM) | 79.2% | — |
Multimodal Not comparable
GPT-5: 46.8 (#13), Inkling: —
| Benchmark | GPT-5 | Inkling |
|---|---|---|
| LMArena Vision | 1232 | — |
| GeoBench | 81% | — |
| VPCT | 66% | — |
Multilingual Inkling leads
GPT-5: 51.4 (#110), Inkling: 54.0 (#52)
| Benchmark | GPT-5 | Inkling |
|---|---|---|
| LMArena Non-English | 1397 | 1434 |
| LMArena Chinese | 1422 | 1490 |
| LMArena French | 1410 | 1458 |
| LMArena German | 1416 | 1446 |
| LMArena Japanese | 1409 | 1429 |
| LMArena Korean | 1360 | 1404 |
| LMArena Russian | 1406 | 1429 |
| LMArena Spanish | 1399 | 1448 |
Instruction Following Inkling leads
GPT-5: 73.8 (#113), Inkling: 75.1 (#71)
| Benchmark | GPT-5 | Inkling |
|---|---|---|
| LMArena Instruction Following | 1388 | 1426 |
| IFEval | 87.5% | — |
Long Context GPT-5 leads
GPT-5: 69.5 (#2), Inkling: 43.8 (#86)
| Benchmark | GPT-5 | Inkling |
|---|---|---|
| LMArena Longer Query | 1399 | 1434 |
| Fiction.LiveBench | 97.2% | — |
Writing & Preference Inkling leads
GPT-5: 63.4 (#65), Inkling: 65.2 (#51)
| Benchmark | GPT-5 | Inkling |
|---|---|---|
| LMArena Text | 1406 | 1441 |
| LMArena Creative Writing | 1365 | 1387 |
| EQ-Bench Creative Writing | 1627 | 1611 |
| LMArena Multi-Turn | 1426 | 1436 |
| Short-Story Creative Writing | 86% | — |
| WildBench | 85.7% | — |
| EQ-Bench 4 | — | 1226 |
Frequently asked questions
Is GPT-5 better than Inkling?
GPT-5 is the stronger model overall, scoring 50.9 to 44.1 on the Noometry Index.
Which is cheaper, GPT-5 or Inkling?
Inkling is cheaper. It lists at $1.87 per million input tokens and $4.68 per million output tokens; GPT-5 lists at $1.25 and $10.
Is GPT-5 or Inkling better for coding?
GPT-5 scores higher on coding benchmarks: 50.3 versus 34.5 in the Noometry coding category.
Which has the bigger context window?
GPT-5 does, with 400K tokens against 66K.
How many benchmarks do GPT-5 and Inkling share?
36 benchmarks have published results for both models. GPT-5 has 69 scored results on Noometry and Inkling has 41.