Model comparison
GPT-5.2 Codex vs Grok 3
GPT-5.2 Codex is the stronger model overall, scoring 42.6 to 39.9 on the Noometry Index.
Last verified . 0 shared benchmarks.
Summary
- The widest gap is in agentic & tool use, where GPT-5.2 Codex leads 41.0 to 30.5.
Side by side
| GPT-5.2 Codex | Grok 3 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 42.6 | 39.9 |
| Released | 2025-12-18 | 2025-04-09 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $1.75 | — |
| Output $ / M tokens | $14 | — |
| Results tracked | 5 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.2 Codex leads
GPT-5.2 Codex: 45.5 (#71), Grok 3: 41.9 (#115)
| Benchmark | GPT-5.2 Codex | Grok 3 |
|---|---|---|
| SWE-bench Verified (bash only) | 72.8% | — |
| Aider Polyglot | — | 53.3% |
| LMArena WebDev | 1339 | — |
| SWE-bench Multilingual | 66.3% | — |
| WeirdML | — | 37.2% |
| LMArena Coding | — | 1432 |
| ALE-Bench | 1,300 | — |
Agentic & Tool Use GPT-5.2 Codex leads
GPT-5.2 Codex: 41.0 (#22), Grok 3: 30.5 (#76)
| Benchmark | GPT-5.2 Codex | Grok 3 |
|---|---|---|
| Terminal-Bench | 66.5% | — |
| BALROG | — | 29.5% |
Reasoning Not comparable
GPT-5.2 Codex: —, Grok 3: 13.7 (#333)
| Benchmark | GPT-5.2 Codex | Grok 3 |
|---|---|---|
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 36.1% |
| Kagi LLM Benchmark | — | 61.3% |
| ARC-AGI-1 | — | 5.5% |
| LMArena Hard Prompts | — | 1434 |
| Epoch Capabilities Index | — | 138.33 |
Math Not comparable
GPT-5.2 Codex: —, Grok 3: 38.0 (#145)
| Benchmark | GPT-5.2 Codex | Grok 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 55.6% |
| Omni-MATH | — | 46.4% |
| LMArena Math | — | 1391 |
| MATH Level 5 | — | 88.7% |
| FrontierMath (Feb 2025 set) | — | 3.8% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Not comparable
GPT-5.2 Codex: —, Grok 3: 46.2 (#82)
| Benchmark | GPT-5.2 Codex | Grok 3 |
|---|---|---|
| GPQA Diamond | — | 75.8% |
| MMLU-Pro | — | 78.8% |
| Confabulations | — | 14.2% |
| Vectara Hallucination Rate | — | 5.8% |
| GPQA (HELM) | — | 65% |
| LMArena Expert | — | 1421 |
Multilingual Not comparable
GPT-5.2 Codex: —, Grok 3: 52.3 (#87)
| Benchmark | GPT-5.2 Codex | Grok 3 |
|---|---|---|
| LMArena Non-English | — | 1410 |
| LMArena Chinese | — | 1448 |
| LMArena French | — | 1460 |
| LMArena German | — | 1431 |
| LMArena Japanese | — | 1387 |
| LMArena Korean | — | 1373 |
| LMArena Russian | — | 1416 |
| LMArena Spanish | — | 1417 |
Instruction Following Not comparable
GPT-5.2 Codex: —, Grok 3: 75.0 (#73)
| Benchmark | GPT-5.2 Codex | Grok 3 |
|---|---|---|
| IFEval | — | 88.4% |
| LMArena Instruction Following | — | 1409 |
Long Context Not comparable
GPT-5.2 Codex: —, Grok 3: 38.7 (#192)
| Benchmark | GPT-5.2 Codex | Grok 3 |
|---|---|---|
| Fiction.LiveBench | — | 58.3% |
| LMArena Longer Query | — | 1439 |
Writing & Preference Not comparable
GPT-5.2 Codex: —, Grok 3: 55.8 (#141)
| Benchmark | GPT-5.2 Codex | Grok 3 |
|---|---|---|
| LMArena Text | — | 1426 |
| LMArena Creative Writing | — | 1414 |
| Short-Story Creative Writing | — | 76.4% |
| EQ-Bench Creative Writing | — | 1186 |
| WildBench | — | 84.9% |
| LMArena Multi-Turn | — | 1425 |
Frequently asked questions
Is GPT-5.2 Codex better than Grok 3?
GPT-5.2 Codex is the stronger model overall, scoring 42.6 to 39.9 on the Noometry Index.
Is GPT-5.2 Codex or Grok 3 better for coding?
GPT-5.2 Codex scores higher on coding benchmarks: 45.5 versus 41.9 in the Noometry coding category.
How many benchmarks do GPT-5.2 Codex and Grok 3 share?
0 benchmarks have published results for both models. GPT-5.2 Codex has 5 scored results on Noometry and Grok 3 has 40.