Model comparison
GPT-5.4 nano vs Grok 4.3
Grok 4.3 is the stronger model overall, scoring 43.6 to 41.9 on the Noometry Index. GPT-5.4 nano costs 3.4× less per token, which makes it the better buy when Grok 4.3's lead doesn't matter for your workload.
Last verified . 33 shared benchmarks.
Summary
- They share 33 benchmarks with published results for both. GPT-5.4 nano scores higher in 2 categories and Grok 4.3 in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.3 leads 34.9 to 23.8.
- The biggest single-benchmark swing is SimpleQA Verified: 11.7% for GPT-5.4 nano and 33.2% for Grok 4.3.
- GPT-5.4 nano is cheaper at $0.20 / $1.25 per million input/output tokens, against $1.25 / $2.50 for Grok 4.3.
- Grok 4.3 accepts more context: 1M tokens versus 400K.
Side by side
| GPT-5.4 nano | Grok 4.3 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 41.9 | 43.6 |
| Released | 2026-03-17 | 2026-04-17 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 1M |
| Max output | 128K | 30K |
| Input $ / M tokens | $0.20 | $1.25 |
| Output $ / M tokens | $1.25 | $2.50 |
| Results tracked | 40 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 nano leads
GPT-5.4 nano: 43.1 (#82), Grok 4.3: 41.3 (#117)
| Benchmark | GPT-5.4 nano | Grok 4.3 |
|---|---|---|
| SciCode | 46.9% | 47.3% |
| WeirdML | 49.2% | 49.9% |
| LMArena Coding | 1404 | 1415 |
| ALE-Bench | 1,005 | 944.17 |
| LMArena WebDev | — | 1357 |
Agentic & Tool Use Not comparable
GPT-5.4 nano: —, Grok 4.3: 27.6 (#93)
| Benchmark | GPT-5.4 nano | Grok 4.3 |
|---|---|---|
| GDP.pdf | — | 8% |
| LMArena Search | — | 1165 |
| Vending-Bench 2 | — | 35.26 |
Reasoning Grok 4.3 leads
GPT-5.4 nano: 23.8 (#159), Grok 4.3: 34.9 (#60)
| Benchmark | GPT-5.4 nano | Grok 4.3 |
|---|---|---|
| CritPt | 9.3% | 8% |
| Chess Puzzles | 30% | 25% |
| LMArena Hard Prompts | 1380 | 1397 |
| DTBench | 80.3% | 90.7% |
| LMCA | 36.9% | 38.3% |
| Epoch Capabilities Index | 145.78 | 149.13 |
| ForecastBench | 57.3 | 60.3 |
| ARC-AGI-2 | 5.7% | — |
| Kagi LLM Benchmark | 39.7% | — |
| NYT Connections (extended) | — | 55.2% |
| ARC-AGI-1 | 51.5% | — |
| Mystery Game Puzzles | 9% | — |
Math Grok 4.3 leads
GPT-5.4 nano: 40.7 (#76), Grok 4.3: 45.8 (#60)
| Benchmark | GPT-5.4 nano | Grok 4.3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 44.9% | 42.8% |
| FrontierMath Tier 4 | 12.2% | 14.6% |
| OTIS Mock AIME 2024-2025 | 87.8% | 93.3% |
| ProofBench | 5% | 11% |
| LMArena Math | 1406 | 1388 |
| FrontierMath (Feb 2025 set) | 25.9% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge Grok 4.3 leads
GPT-5.4 nano: 41.8 (#94), Grok 4.3: 52.5 (#57)
| Benchmark | GPT-5.4 nano | Grok 4.3 |
|---|---|---|
| GPQA Diamond | 78.5% | 88.8% |
| SimpleQA Verified | 11.7% | 33.2% |
| LMArena Expert | 1397 | 1385 |
| Vectara Hallucination Rate | 3.1% | — |
Multimodal GPT-5.4 nano leads
GPT-5.4 nano: 36.0 (#73), Grok 4.3: 31.6 (#96)
| Benchmark | GPT-5.4 nano | Grok 4.3 |
|---|---|---|
| LMArena Vision | 1196 | 1231 |
| Blueprint-Bench 2 | — | 0% |
Multilingual Grok 4.3 leads
GPT-5.4 nano: 48.8 (#136), Grok 4.3: 50.7 (#114)
| Benchmark | GPT-5.4 nano | Grok 4.3 |
|---|---|---|
| LMArena Non-English | 1359 | 1385 |
| LMArena Chinese | 1391 | 1423 |
| LMArena French | 1396 | 1412 |
| LMArena German | 1368 | 1395 |
| LMArena Japanese | 1342 | 1379 |
| LMArena Korean | 1321 | 1357 |
| LMArena Russian | 1363 | 1399 |
| LMArena Spanish | 1368 | 1402 |
Instruction Following Too close to call
GPT-5.4 nano: 72.1 (#139), Grok 4.3: 72.3 (#135)
| Benchmark | GPT-5.4 nano | Grok 4.3 |
|---|---|---|
| LMArena Instruction Following | 1362 | 1366 |
Long Context Too close to call
GPT-5.4 nano: 41.6 (#131), Grok 4.3: 42.5 (#118)
| Benchmark | GPT-5.4 nano | Grok 4.3 |
|---|---|---|
| LMArena Longer Query | 1365 | 1393 |
Writing & Preference Grok 4.3 leads
GPT-5.4 nano: 55.9 (#137), Grok 4.3: 58.6 (#112)
| Benchmark | GPT-5.4 nano | Grok 4.3 |
|---|---|---|
| LMArena Text | 1372 | 1397 |
| LMArena Creative Writing | 1313 | 1380 |
| LMArena Multi-Turn | 1382 | 1406 |
| EQ-Bench 4 | — | 1075 |
Frequently asked questions
Is GPT-5.4 nano better than Grok 4.3?
Grok 4.3 is the stronger model overall, scoring 43.6 to 41.9 on the Noometry Index. GPT-5.4 nano costs 3.4× less per token, which makes it the better buy when Grok 4.3's lead doesn't matter for your workload.
Which is cheaper, GPT-5.4 nano or Grok 4.3?
GPT-5.4 nano is cheaper. It lists at $0.20 per million input tokens and $1.25 per million output tokens; Grok 4.3 lists at $1.25 and $2.50.
Is GPT-5.4 nano or Grok 4.3 better for coding?
GPT-5.4 nano scores higher on coding benchmarks: 43.1 versus 41.3 in the Noometry coding category.
Which has the bigger context window?
Grok 4.3 does, with 1M tokens against 400K.
How many benchmarks do GPT-5.4 nano and Grok 4.3 share?
33 benchmarks have published results for both models. GPT-5.4 nano has 40 scored results on Noometry and Grok 4.3 has 40.