Model comparison
gpt-oss-120b vs Grok 4.3
Grok 4.3 is the stronger model overall, scoring 43.8 to 36.3 on the Noometry Index. gpt-oss-120b costs 22× less per token, which makes it the better buy when Grok 4.3's lead doesn't matter for your workload.
Last verified . 28 shared benchmarks.
Summary
- They share 28 benchmarks with published results for both. gpt-oss-120b scores higher in 1 category and Grok 4.3 in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.3 leads 35.9 to 20.0.
- The biggest single-benchmark swing is LMCA: 22.1% for gpt-oss-120b and 38.3% for Grok 4.3.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $1.25 / $2.50 for Grok 4.3.
- Grok 4.3 accepts more context: 1M tokens versus 131K.
- gpt-oss-120b has downloadable open weights; the other is API-only.
Side by side
| gpt-oss-120b | Grok 4.3 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 36.3 | 43.8 |
| Released | 2025-08-05 | 2026-04-17 |
| Weights | Open | Proprietary |
| Context window | 131K | 1M |
| Max output | 41K | 30K |
| Input $ / M tokens | $0.037 | $1.25 |
| Output $ / M tokens | $0.17 | $2.50 |
| Results tracked | 48 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.3 leads
gpt-oss-120b: 33.5 (#256), Grok 4.3: 41.6 (#121)
| Benchmark | gpt-oss-120b | Grok 4.3 |
|---|---|---|
| SciCode | 36% | 47.3% |
| WeirdML | 48.2% | 49.9% |
| LMArena Coding | 1380 | 1415 |
| ALE-Bench | 575.62 | 944.17 |
| SWE-bench Verified (bash only) | 26% | — |
| Aider Polyglot | 41.8% | — |
| LMArena WebDev | — | 1357 |
| AlgoTune | 1.41 | — |
Agentic & Tool Use Grok 4.3 leads
gpt-oss-120b: 12.2 (#153), Grok 4.3: 27.7 (#99)
| Benchmark | gpt-oss-120b | Grok 4.3 |
|---|---|---|
| Vending-Bench 2 | -21.53 | 35.26 |
| Terminal-Bench | 18.7% | — |
| APEX-Agents | 4.4% | — |
| GDP.pdf | — | 8% |
| LMArena Search | — | 1165 |
| METR Time Horizons | 56.6% | — |
Reasoning Grok 4.3 leads
gpt-oss-120b: 20.0 (#245), Grok 4.3: 35.9 (#68)
| Benchmark | gpt-oss-120b | Grok 4.3 |
|---|---|---|
| CritPt | 1.1% | 8% |
| Chess Puzzles | 20% | 25% |
| LMArena Hard Prompts | 1364 | 1396 |
| DTBench | 76.3% | 90.7% |
| LMCA | 22.1% | 38.3% |
| Epoch Capabilities Index | 139.93 | 149.16 |
| SimpleBench | 22.1% | — |
| Kagi LLM Benchmark | 58.6% | — |
| NYT Connections (extended) | — | 55.2% |
| Mystery Game Puzzles | 2% | — |
| Surface Evolver Bench | 25% | — |
| ForecastBench | — | 60.3 |
Math gpt-oss-120b leads
gpt-oss-120b: 52.5 (#50), Grok 4.3: 46.0 (#74)
| Benchmark | gpt-oss-120b | Grok 4.3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 93.3% |
| LMArena Math | 1389 | 1388 |
| FrontierMath (Tiers 1-3) | — | 42.8% |
| FrontierMath Tier 4 | — | 14.6% |
| ProofBench | — | 11% |
| Omni-MATH | 68.8% | — |
Knowledge Grok 4.3 leads
gpt-oss-120b: 42.4 (#96), Grok 4.3: 52.5 (#62)
| Benchmark | gpt-oss-120b | Grok 4.3 |
|---|---|---|
| GPQA Diamond | 75.8% | 88.8% |
| LMArena Expert | 1356 | 1385 |
| SimpleQA Verified | — | 33.2% |
| MMLU-Pro | 79.5% | — |
| Confabulations | 15.7% | — |
| Vectara Hallucination Rate | 14.2% | — |
| GPQA (HELM) | 68.4% | — |
Multimodal Not comparable
gpt-oss-120b: —, Grok 4.3: 31.6 (#104)
| Benchmark | gpt-oss-120b | Grok 4.3 |
|---|---|---|
| LMArena Vision | — | 1229 |
| Blueprint-Bench 2 | — | 0% |
Multilingual Grok 4.3 leads
gpt-oss-120b: 48.0 (#147), Grok 4.3: 50.5 (#120)
| Benchmark | gpt-oss-120b | Grok 4.3 |
|---|---|---|
| LMArena Non-English | 1351 | 1385 |
| LMArena Chinese | 1385 | 1422 |
| LMArena French | 1369 | 1412 |
| LMArena German | 1353 | 1395 |
| LMArena Japanese | 1331 | 1379 |
| LMArena Korean | 1282 | 1356 |
| LMArena Russian | 1343 | 1399 |
| LMArena Spanish | 1389 | 1398 |
Instruction Following Grok 4.3 leads
gpt-oss-120b: 69.3 (#173), Grok 4.3: 72.1 (#140)
| Benchmark | gpt-oss-120b | Grok 4.3 |
|---|---|---|
| LMArena Instruction Following | 1318 | 1366 |
| IFEval | 83.6% | — |
Long Context Grok 4.3 leads
gpt-oss-120b: 31.4 (#278), Grok 4.3: 42.5 (#123)
| Benchmark | gpt-oss-120b | Grok 4.3 |
|---|---|---|
| LMArena Longer Query | 1319 | 1393 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference Grok 4.3 leads
gpt-oss-120b: 46.5 (#217), Grok 4.3: 58.5 (#118)
| Benchmark | gpt-oss-120b | Grok 4.3 |
|---|---|---|
| LMArena Text | 1365 | 1397 |
| LMArena Creative Writing | 1275 | 1380 |
| LMArena Multi-Turn | 1340 | 1406 |
| Short-Story Creative Writing | 77.1% | — |
| EQ-Bench Creative Writing | 961 | — |
| WildBench | 84.5% | — |
| EQ-Bench 4 | — | 1075 |
Frequently asked questions
Is gpt-oss-120b better than Grok 4.3?
Grok 4.3 is the stronger model overall, scoring 43.8 to 36.3 on the Noometry Index. gpt-oss-120b costs 22× 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-oss-120b or Grok 4.3?
gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; Grok 4.3 lists at $1.25 and $2.50.
Is gpt-oss-120b or Grok 4.3 better for coding?
Grok 4.3 scores higher on coding benchmarks: 41.6 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
Grok 4.3 does, with 1M tokens against 131K.
How many benchmarks do gpt-oss-120b and Grok 4.3 share?
28 benchmarks have published results for both models. gpt-oss-120b has 48 scored results on Noometry and Grok 4.3 has 40.