Model comparison
GPT-5.4 nano vs Grok 4.7
Grok 4.7 is the stronger model overall, scoring 46.6 to 41.9 on the Noometry Index. GPT-5.4 nano costs 6.5× less per token, which makes it the better buy when Grok 4.7's lead doesn't matter for your workload.
Last verified . 15 shared benchmarks.
Summary
- They share 15 benchmarks with published results for both. GPT-5.4 nano scores higher in 2 categories and Grok 4.7 in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.7 leads 40.1 to 23.8.
- The biggest single-benchmark swing is ProofBench: 5% for GPT-5.4 nano and 34% for Grok 4.7.
- GPT-5.4 nano is cheaper at $0.20 / $1.25 per million input/output tokens, against $2 / $6 for Grok 4.7.
- Grok 4.7 accepts more context: 500K tokens versus 400K.
Side by side
| GPT-5.4 nano | Grok 4.7 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 41.9 | 46.6 |
| Released | 2026-03-17 | 2026-09-21 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 500K |
| Max output | 128K | 500K |
| Input $ / M tokens | $0.20 | $2 |
| Output $ / M tokens | $1.25 | $6 |
| Results tracked | 40 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.7 leads
GPT-5.4 nano: 43.1 (#82), Grok 4.7: 55.0 (#23)
| Benchmark | GPT-5.4 nano | Grok 4.7 |
|---|---|---|
| SciCode | 46.9% | 57.4% |
| LMArena Coding | 1404 | 1423 |
| CursorBench | — | 46.3% |
| LMArena WebDev | — | 1636 |
| FrontierSWE | — | 29.5% |
| WeirdML | 49.2% | — |
| ALE-Bench | 1,005 | — |
Reasoning Grok 4.7 leads
GPT-5.4 nano: 23.8 (#159), Grok 4.7: 40.1 (#49)
| Benchmark | GPT-5.4 nano | Grok 4.7 |
|---|---|---|
| CritPt | 9.3% | 17.7% |
| LMArena Hard Prompts | 1380 | 1409 |
| ARC-AGI-2 | 5.7% | — |
| Kagi LLM Benchmark | 39.7% | — |
| NYT Connections (extended) | — | 76.8% |
| ARC-AGI-1 | 51.5% | — |
| Chess Puzzles | 30% | — |
| Mystery Game Puzzles | 9% | — |
| DTBench | 80.3% | — |
| LMCA | 36.9% | — |
| Epoch Capabilities Index | 145.78 | — |
| ForecastBench | 57.3 | — |
Math Too close to call
GPT-5.4 nano: 40.7 (#76), Grok 4.7: 41.1 (#73)
| Benchmark | GPT-5.4 nano | Grok 4.7 |
|---|---|---|
| ProofBench | 5% | 34% |
| LMArena Math | 1406 | 1393 |
| FrontierMath (Tiers 1-3) | 44.9% | — |
| FrontierMath Tier 4 | 12.2% | — |
| OTIS Mock AIME 2024-2025 | 87.8% | — |
| FrontierMath (Feb 2025 set) | 25.9% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge GPT-5.4 nano leads
GPT-5.4 nano: 41.8 (#94), Grok 4.7: 38.7 (#130)
| Benchmark | GPT-5.4 nano | Grok 4.7 |
|---|---|---|
| LMArena Expert | 1397 | 1413 |
| GPQA Diamond | 78.5% | — |
| SimpleQA Verified | 11.7% | — |
| Vectara Hallucination Rate | 3.1% | — |
Multimodal GPT-5.4 nano leads
GPT-5.4 nano: 36.0 (#73), Grok 4.7: 28.8 (#107)
| Benchmark | GPT-5.4 nano | Grok 4.7 |
|---|---|---|
| LMArena Vision | 1196 | — |
| Furniture Assembly | — | 20.8% |
Multilingual Grok 4.7 leads
GPT-5.4 nano: 48.8 (#136), Grok 4.7: 51.2 (#108)
| Benchmark | GPT-5.4 nano | Grok 4.7 |
|---|---|---|
| LMArena Non-English | 1359 | 1392 |
| LMArena Chinese | 1391 | 1476 |
| LMArena Russian | 1363 | 1396 |
| LMArena French | 1396 | — |
| LMArena German | 1368 | — |
| LMArena Japanese | 1342 | — |
| LMArena Korean | 1321 | — |
| LMArena Spanish | 1368 | — |
Instruction Following Grok 4.7 leads
GPT-5.4 nano: 72.1 (#139), Grok 4.7: 74.3 (#98)
| Benchmark | GPT-5.4 nano | Grok 4.7 |
|---|---|---|
| LMArena Instruction Following | 1362 | 1406 |
Long Context Grok 4.7 leads
GPT-5.4 nano: 41.6 (#131), Grok 4.7: 43.2 (#97)
| Benchmark | GPT-5.4 nano | Grok 4.7 |
|---|---|---|
| LMArena Longer Query | 1365 | 1416 |
Writing & Preference Grok 4.7 leads
GPT-5.4 nano: 55.9 (#137), Grok 4.7: 70.1 (#23)
| Benchmark | GPT-5.4 nano | Grok 4.7 |
|---|---|---|
| LMArena Text | 1372 | 1397 |
| LMArena Creative Writing | 1313 | 1391 |
| LMArena Multi-Turn | 1382 | 1389 |
| EQ-Bench Creative Writing | — | 2007 |
Frequently asked questions
Is GPT-5.4 nano better than Grok 4.7?
Grok 4.7 is the stronger model overall, scoring 46.6 to 41.9 on the Noometry Index. GPT-5.4 nano costs 6.5× less per token, which makes it the better buy when Grok 4.7's lead doesn't matter for your workload.
Which is cheaper, GPT-5.4 nano or Grok 4.7?
GPT-5.4 nano is cheaper. It lists at $0.20 per million input tokens and $1.25 per million output tokens; Grok 4.7 lists at $2 and $6.
Is GPT-5.4 nano or Grok 4.7 better for coding?
Grok 4.7 scores higher on coding benchmarks: 55.0 versus 43.1 in the Noometry coding category.
Which has the bigger context window?
Grok 4.7 does, with 500K tokens against 400K.
How many benchmarks do GPT-5.4 nano and Grok 4.7 share?
15 benchmarks have published results for both models. GPT-5.4 nano has 40 scored results on Noometry and Grok 4.7 has 21.