Model comparison
Grok 4.20 Multi-Agent vs Llama 3.2 90B
Grok 4.20 Multi-Agent is the stronger model overall, scoring 45.5 to 27.1 on the Noometry Index.
Last verified . 1 shared benchmarks.
MetaSummary
- They share 1 benchmark with published results for both. Grok 4.20 Multi-Agent scores higher in 4 categories and Llama 3.2 90B in 0 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 4.20 Multi-Agent leads 36.6 to 10.4.
- Llama 3.2 90B has downloadable open weights; the other is API-only.
Side by side
| Grok 4.20 Multi-Agent | Llama 3.2 90B | |
|---|---|---|
| Provider | xAI | Meta |
| Noometry Index | 45.5 | 27.1 |
| Released | 2026-03-09 | 2024-09-24 |
| Weights | Proprietary | Open |
| Context window | 1M | — |
| Max output | 30K | — |
| Input $ / M tokens | $1.25 | — |
| Output $ / M tokens | $2.50 | — |
| Results tracked | 20 | 9 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Grok 4.20 Multi-Agent: 42.4 (#91), Llama 3.2 90B: —
| Benchmark | Grok 4.20 Multi-Agent | Llama 3.2 90B |
|---|---|---|
| LMArena Coding | 1458 | — |
Agentic & Tool Use Not comparable
Grok 4.20 Multi-Agent: —, Llama 3.2 90B: 29.7 (#73)
| Benchmark | Grok 4.20 Multi-Agent | Llama 3.2 90B |
|---|---|---|
| BALROG | — | 27.3% |
| LMArena Search | 1204 | — |
Reasoning Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 42.4 (#44), Llama 3.2 90B: 20.9 (#207)
| Benchmark | Grok 4.20 Multi-Agent | Llama 3.2 90B |
|---|---|---|
| NYT Connections (extended) | 89.6% | — |
| EnigmaEval | — | 0.4% |
| LMArena Hard Prompts | 1448 | — |
| Epoch Capabilities Index | — | 125.5 |
Math Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 36.6 (#106), Llama 3.2 90B: 10.4 (#302)
| Benchmark | Grok 4.20 Multi-Agent | Llama 3.2 90B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 2.6% |
| LMArena Math | 1441 | — |
| MATH Level 5 | — | 39.4% |
Knowledge Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 39.7 (#116), Llama 3.2 90B: 21.7 (#269)
| Benchmark | Grok 4.20 Multi-Agent | Llama 3.2 90B |
|---|---|---|
| GPQA Diamond | — | 41% |
| LMArena Expert | 1445 | — |
| MMLU | — | 80.3% |
Multimodal Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 39.8 (#44), Llama 3.2 90B: 24.9 (#119)
| Benchmark | Grok 4.20 Multi-Agent | Llama 3.2 90B |
|---|---|---|
| LMArena Vision | 1259 | 1000 |
| GeoBench | — | 52% |
Multilingual Not comparable
Grok 4.20 Multi-Agent: 54.6 (#40), Llama 3.2 90B: —
| Benchmark | Grok 4.20 Multi-Agent | Llama 3.2 90B |
|---|---|---|
| LMArena Non-English | 1440 | — |
| LMArena Chinese | 1475 | — |
| LMArena French | 1467 | — |
| LMArena German | 1456 | — |
| LMArena Japanese | 1405 | — |
| LMArena Korean | 1416 | — |
| LMArena Russian | 1457 | — |
| LMArena Spanish | 1447 | — |
Instruction Following Not comparable
Grok 4.20 Multi-Agent: 75.0 (#80), Llama 3.2 90B: —
| Benchmark | Grok 4.20 Multi-Agent | Llama 3.2 90B |
|---|---|---|
| LMArena Instruction Following | 1420 | — |
Long Context Not comparable
Grok 4.20 Multi-Agent: 43.7 (#84), Llama 3.2 90B: —
| Benchmark | Grok 4.20 Multi-Agent | Llama 3.2 90B |
|---|---|---|
| LMArena Longer Query | 1430 | — |
Writing & Preference Not comparable
Grok 4.20 Multi-Agent: 64.1 (#57), Llama 3.2 90B: —
| Benchmark | Grok 4.20 Multi-Agent | Llama 3.2 90B |
|---|---|---|
| LMArena Text | 1450 | — |
| LMArena Creative Writing | 1435 | — |
| LMArena Multi-Turn | 1451 | — |
Frequently asked questions
Is Grok 4.20 Multi-Agent better than Llama 3.2 90B?
Grok 4.20 Multi-Agent is the stronger model overall, scoring 45.5 to 27.1 on the Noometry Index.
How many benchmarks do Grok 4.20 Multi-Agent and Llama 3.2 90B share?
1 benchmark has published results for both models. Grok 4.20 Multi-Agent has 20 scored results on Noometry and Llama 3.2 90B has 9.