Model comparison
gpt-oss-120b vs Grok 4.20 Multi-Agent
Grok 4.20 Multi-Agent is the stronger model overall, scoring 45.5 to 36.0 on the Noometry Index. gpt-oss-120b costs 22× less per token, which makes it the better buy when Grok 4.20 Multi-Agent's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
Summary
- They share 17 benchmarks with published results for both. gpt-oss-120b scores higher in 2 categories and Grok 4.20 Multi-Agent in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.20 Multi-Agent leads 42.4 to 19.5.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $1.25 / $2.50 for Grok 4.20 Multi-Agent.
- Grok 4.20 Multi-Agent 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 4.20 Multi-Agent | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 36.0 | 45.5 |
| Released | 2025-08-05 | 2026-03-09 |
| 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 | 20 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.20 Multi-Agent leads
gpt-oss-120b: 33.2 (#251), Grok 4.20 Multi-Agent: 42.4 (#91)
| Benchmark | gpt-oss-120b | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Coding | 1380 | 1458 |
| SWE-bench Verified (bash only) | 26% | — |
| Aider Polyglot | 41.8% | — |
| SciCode | 36% | — |
| WeirdML | 48.2% | — |
| ALE-Bench | 575.62 | — |
| AlgoTune | 1.41 | — |
Agentic & Tool Use Not comparable
gpt-oss-120b: 12.1 (#145), Grok 4.20 Multi-Agent: —
| Benchmark | gpt-oss-120b | Grok 4.20 Multi-Agent |
|---|---|---|
| Terminal-Bench | 18.7% | — |
| APEX-Agents | 4.4% | — |
| LMArena Search | — | 1204 |
| METR Time Horizons | 56.6% | — |
| Vending-Bench 2 | -21.53 | — |
Reasoning Grok 4.20 Multi-Agent leads
gpt-oss-120b: 19.5 (#236), Grok 4.20 Multi-Agent: 42.4 (#44)
| Benchmark | gpt-oss-120b | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Hard Prompts | 1364 | 1448 |
| SimpleBench | 22.1% | — |
| Kagi LLM Benchmark | 58.6% | — |
| NYT Connections (extended) | — | 89.6% |
| CritPt | 1.1% | — |
| Chess Puzzles | 20% | — |
| Mystery Game Puzzles | 2% | — |
| DTBench | 76.3% | — |
| LMCA | 22.1% | — |
| Surface Evolver Bench | 25% | — |
| Epoch Capabilities Index | 139.94 | — |
Math gpt-oss-120b leads
gpt-oss-120b: 50.6 (#43), Grok 4.20 Multi-Agent: 36.6 (#106)
| Benchmark | gpt-oss-120b | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Math | 1388 | 1441 |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| Omni-MATH | 68.8% | — |
Knowledge gpt-oss-120b leads
gpt-oss-120b: 42.3 (#89), Grok 4.20 Multi-Agent: 39.7 (#116)
| Benchmark | gpt-oss-120b | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Expert | 1356 | 1445 |
| GPQA Diamond | 75.8% | — |
| MMLU-Pro | 79.5% | — |
| Confabulations | 15.7% | — |
| Vectara Hallucination Rate | 14.2% | — |
| GPQA (HELM) | 68.4% | — |
Multimodal Not comparable
gpt-oss-120b: —, Grok 4.20 Multi-Agent: 39.8 (#44)
| Benchmark | gpt-oss-120b | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Vision | — | 1259 |
Multilingual Grok 4.20 Multi-Agent leads
gpt-oss-120b: 48.2 (#142), Grok 4.20 Multi-Agent: 54.6 (#40)
| Benchmark | gpt-oss-120b | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Non-English | 1350 | 1440 |
| LMArena Chinese | 1384 | 1475 |
| LMArena French | 1370 | 1467 |
| LMArena German | 1355 | 1456 |
| LMArena Japanese | 1332 | 1405 |
| LMArena Korean | 1283 | 1416 |
| LMArena Russian | 1343 | 1457 |
| LMArena Spanish | 1388 | 1447 |
Instruction Following Grok 4.20 Multi-Agent leads
gpt-oss-120b: 69.5 (#168), Grok 4.20 Multi-Agent: 75.0 (#80)
| Benchmark | gpt-oss-120b | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Instruction Following | 1318 | 1420 |
| IFEval | 83.6% | — |
Long Context Grok 4.20 Multi-Agent leads
gpt-oss-120b: 31.4 (#273), Grok 4.20 Multi-Agent: 43.7 (#84)
| Benchmark | gpt-oss-120b | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Longer Query | 1319 | 1430 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference Grok 4.20 Multi-Agent leads
gpt-oss-120b: 46.7 (#213), Grok 4.20 Multi-Agent: 64.1 (#57)
| Benchmark | gpt-oss-120b | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Text | 1366 | 1450 |
| LMArena Creative Writing | 1274 | 1435 |
| LMArena Multi-Turn | 1340 | 1451 |
| Short-Story Creative Writing | 77.1% | — |
| EQ-Bench Creative Writing | 961 | — |
| WildBench | 84.5% | — |
Frequently asked questions
Is gpt-oss-120b better than Grok 4.20 Multi-Agent?
Grok 4.20 Multi-Agent is the stronger model overall, scoring 45.5 to 36.0 on the Noometry Index. gpt-oss-120b costs 22× less per token, which makes it the better buy when Grok 4.20 Multi-Agent's lead doesn't matter for your workload.
Which is cheaper, gpt-oss-120b or Grok 4.20 Multi-Agent?
gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; Grok 4.20 Multi-Agent lists at $1.25 and $2.50.
Is gpt-oss-120b or Grok 4.20 Multi-Agent better for coding?
Grok 4.20 Multi-Agent scores higher on coding benchmarks: 42.4 versus 33.2 in the Noometry coding category.
Which has the bigger context window?
Grok 4.20 Multi-Agent does, with 1M tokens against 131K.
How many benchmarks do gpt-oss-120b and Grok 4.20 Multi-Agent share?
17 benchmarks have published results for both models. gpt-oss-120b has 48 scored results on Noometry and Grok 4.20 Multi-Agent has 20.