Model comparison
GPT-5.3 Codex vs Grok 4 Fast
GPT-5.3 Codex is the stronger model overall, scoring 45.8 to 39.4 on the Noometry Index.
Last verified . 3 shared benchmarks.
Summary
- They share 3 benchmarks with published results for both. GPT-5.3 Codex scores higher in 2 categories and Grok 4 Fast in 0 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where GPT-5.3 Codex leads 48.0 to 29.5.
- The biggest single-benchmark swing is WeirdML: 79.3% for GPT-5.3 Codex and 42.9% for Grok 4 Fast.
Side by side
| GPT-5.3 Codex | Grok 4 Fast | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 45.8 | 39.4 |
| Released | 2026-02-05 | 2025-09-19 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $1.75 | — |
| Output $ / M tokens | $14 | — |
| Results tracked | 8 | 30 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.3 Codex leads
GPT-5.3 Codex: 48.6 (#56), Grok 4 Fast: 32.5 (#271)
| Benchmark | GPT-5.3 Codex | Grok 4 Fast |
|---|---|---|
| LMArena WebDev | 1409 | 1159 |
| WeirdML | 79.3% | 42.9% |
| SWE-bench Verified | 74.8% | — |
| LMArena Coding | — | 1429 |
| ALE-Bench | 1,655 | — |
Agentic & Tool Use GPT-5.3 Codex leads
GPT-5.3 Codex: 48.0 (#9), Grok 4 Fast: 29.5 (#86)
| Benchmark | GPT-5.3 Codex | Grok 4 Fast |
|---|---|---|
| Terminal-Bench | 78.4% | — |
| τ²-bench Banking | — | 15.7% |
| Cybench | — | 30% |
| LMArena Search | — | 1171 |
| METR Time Horizons | 74.5% | — |
| Vending-Bench 2 | 5,940 | — |
Reasoning Not comparable
GPT-5.3 Codex: —, Grok 4 Fast: 22.2 (#201)
| Benchmark | GPT-5.3 Codex | Grok 4 Fast |
|---|---|---|
| Epoch Capabilities Index | 156.77 | 144.2 |
| ARC-AGI-2 | — | 5.3% |
| Kagi LLM Benchmark | — | 66.1% |
| ARC-AGI-1 | — | 48.5% |
| LMArena Hard Prompts | — | 1412 |
| DTBench | — | 82.7% |
| ForecastBench | — | 60.5 |
Math Not comparable
GPT-5.3 Codex: —, Grok 4 Fast: 38.9 (#123)
| Benchmark | GPT-5.3 Codex | Grok 4 Fast |
|---|---|---|
| LMArena Math | — | 1419 |
Knowledge Not comparable
GPT-5.3 Codex: —, Grok 4 Fast: 32.0 (#214)
| Benchmark | GPT-5.3 Codex | Grok 4 Fast |
|---|---|---|
| Vectara Hallucination Rate | — | 19.7% |
| LMArena Expert | — | 1411 |
Multilingual Not comparable
GPT-5.3 Codex: —, Grok 4 Fast: 51.3 (#111)
| Benchmark | GPT-5.3 Codex | Grok 4 Fast |
|---|---|---|
| LMArena Non-English | — | 1396 |
| LMArena Chinese | — | 1457 |
| LMArena French | — | 1431 |
| LMArena German | — | 1383 |
| LMArena Japanese | — | 1352 |
| LMArena Korean | — | 1357 |
| LMArena Russian | — | 1389 |
| LMArena Spanish | — | 1415 |
Instruction Following Not comparable
GPT-5.3 Codex: —, Grok 4 Fast: 73.2 (#121)
| Benchmark | GPT-5.3 Codex | Grok 4 Fast |
|---|---|---|
| LMArena Instruction Following | — | 1387 |
Long Context Not comparable
GPT-5.3 Codex: —, Grok 4 Fast: 63.2 (#3)
| Benchmark | GPT-5.3 Codex | Grok 4 Fast |
|---|---|---|
| Fiction.LiveBench | — | 94.4% |
| LMArena Longer Query | — | 1415 |
Writing & Preference Not comparable
GPT-5.3 Codex: —, Grok 4 Fast: 60.0 (#102)
| Benchmark | GPT-5.3 Codex | Grok 4 Fast |
|---|---|---|
| LMArena Text | — | 1407 |
| LMArena Creative Writing | — | 1387 |
| LMArena Multi-Turn | — | 1414 |
Frequently asked questions
Is GPT-5.3 Codex better than Grok 4 Fast?
GPT-5.3 Codex is the stronger model overall, scoring 45.8 to 39.4 on the Noometry Index.
Is GPT-5.3 Codex or Grok 4 Fast better for coding?
GPT-5.3 Codex scores higher on coding benchmarks: 48.6 versus 32.5 in the Noometry coding category.
How many benchmarks do GPT-5.3 Codex and Grok 4 Fast share?
3 benchmarks have published results for both models. GPT-5.3 Codex has 8 scored results on Noometry and Grok 4 Fast has 30.