Model comparison
Gemini 2.5 Pro vs Grok 4.20 Multi-Agent
Grok 4.20 Multi-Agent is the stronger model overall, scoring 46.2 to 45.0 on the Noometry Index.
Last verified . 19 shared benchmarks.
GoogleSummary
- They share 19 benchmarks with published results for both. Gemini 2.5 Pro scores higher in 5 categories and Grok 4.20 Multi-Agent in 4 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in long context, where Gemini 2.5 Pro leads 59.8 to 43.7.
- Grok 4.20 Multi-Agent is cheaper at $1.25 / $2.50 per million input/output tokens, against $1.25 / $10 for Gemini 2.5 Pro.
- Gemini 2.5 Pro accepts more context: 1.05M tokens versus 1M.
Side by side
| Gemini 2.5 Pro | Grok 4.20 Multi-Agent | |
|---|---|---|
| Provider | xAI | |
| Noometry Index | 45.0 | 46.2 |
| Released | 2025-03-25 | 2026-03-09 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1M |
| Max output | 66K | 30K |
| Input $ / M tokens | $1.25 | $1.25 |
| Output $ / M tokens | $10 | $2.50 |
| Results tracked | 78 | 20 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Gemini 2.5 Pro: 42.4 (#101), Grok 4.20 Multi-Agent: 43.0 (#92)
| Benchmark | Gemini 2.5 Pro | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Coding | 1452 | 1457 |
| SWE-bench Verified | 57.6% | — |
| SWE-bench Verified (bash only) | 53.6% | — |
| Aider Polyglot | 83.1% | — |
| LMArena WebDev | 1227 | — |
| SciCode | 42.8% | — |
| GSO | 3.9% | — |
| WeirdML | 54% | — |
| LiveBench Coding | 85.9% | — |
| CadEval | 64% | — |
| ALE-Bench | 785.52 | — |
| AlgoTune | 1.51 | — |
Agentic & Tool Use Not comparable
Gemini 2.5 Pro: 29.2 (#88), Grok 4.20 Multi-Agent: —
| Benchmark | Gemini 2.5 Pro | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Search | 1142 | 1204 |
| Terminal-Bench | 32.6% | — |
| GDPval | 23.3% | — |
| Remote Labor Index | 0.8% | — |
| TheAgentCompany | 30.3% | — |
| τ²-bench Banking | 13.7% | — |
| DeepResearch Bench | 42.8% | — |
| BALROG | 43.3% | — |
| METR Time Horizons | 55.4% | — |
| Vending-Bench 2 | 573.64 | — |
Reasoning Grok 4.20 Multi-Agent leads
Gemini 2.5 Pro: 28.8 (#99), Grok 4.20 Multi-Agent: 43.9 (#48)
| Benchmark | Gemini 2.5 Pro | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Hard Prompts | 1455 | 1448 |
| ARC-AGI-2 | 4.9% | — |
| SimpleBench | 62.4% | — |
| Kagi LLM Benchmark | 70.3% | — |
| NYT Connections (extended) | — | 89.6% |
| ARC-AGI-1 | 41% | — |
| CritPt | 2% | — |
| Chess Puzzles | 20% | — |
| EnigmaEval | 5.6% | — |
| LiveBench Reasoning | 89.8% | — |
| DTBench | 82.4% | — |
| LiveBench Data Analysis | 79.9% | — |
| LMCA | 34.8% | — |
| Epoch Capabilities Index | 145.32 | — |
| ForecastBench | 61.3 | — |
| LiveBench | 82.3% | — |
Math Grok 4.20 Multi-Agent leads
Gemini 2.5 Pro: 32.5 (#213), Grok 4.20 Multi-Agent: 39.4 (#104)
| Benchmark | Gemini 2.5 Pro | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Math | 1450 | 1442 |
| FrontierMath (Tiers 1-3) | 24.6% | — |
| FrontierMath Tier 4 | 0% | — |
| OTIS Mock AIME 2024-2025 | 84.7% | — |
| Omni-MATH | 41.6% | — |
| LiveBench Math | 90.2% | — |
| MATH Level 5 | 95.9% | — |
| FrontierMath (Feb 2025 set) | 14.1% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Gemini 2.5 Pro leads
Gemini 2.5 Pro: 56.0 (#46), Grok 4.20 Multi-Agent: 40.4 (#119)
| Benchmark | Gemini 2.5 Pro | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Expert | 1452 | 1445 |
| GPQA Diamond | 85.3% | — |
| Humanity's Last Exam | 21.6% | — |
| MMLU-Pro | 86.3% | — |
| Confabulations | 10.6% | — |
| Vectara Hallucination Rate | 7% | — |
| GPQA (HELM) | 74.9% | — |
Multimodal Gemini 2.5 Pro leads
Gemini 2.5 Pro: 45.2 (#18), Grok 4.20 Multi-Agent: 40.5 (#48)
| Benchmark | Gemini 2.5 Pro | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Vision | 1263 | 1259 |
| GeoBench | 86% | — |
| VPCT | 48% | — |
| LMArena Document | 1421 | — |
| SpatialViz-Bench | 44.7% | — |
Multilingual Too close to call
Gemini 2.5 Pro: 55.3 (#31), Grok 4.20 Multi-Agent: 54.4 (#43)
| Benchmark | Gemini 2.5 Pro | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Non-English | 1451 | 1440 |
| LMArena Chinese | 1507 | 1475 |
| LMArena French | 1472 | 1466 |
| LMArena German | 1487 | 1456 |
| LMArena Japanese | 1461 | 1405 |
| LMArena Korean | 1434 | 1416 |
| LMArena Russian | 1461 | 1457 |
| LMArena Spanish | 1473 | 1447 |
Instruction Following Too close to call
Gemini 2.5 Pro: 75.0 (#75), Grok 4.20 Multi-Agent: 74.8 (#84)
| Benchmark | Gemini 2.5 Pro | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Instruction Following | 1437 | 1420 |
| LiveBench Instruction Following | 80.6% | — |
| IFEval | 84% | — |
Long Context Gemini 2.5 Pro leads
Gemini 2.5 Pro: 59.8 (#5), Grok 4.20 Multi-Agent: 43.7 (#88)
| Benchmark | Gemini 2.5 Pro | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Longer Query | 1449 | 1431 |
| Fiction.LiveBench | 91.7% | — |
Writing & Preference Too close to call
Gemini 2.5 Pro: 63.7 (#62), Grok 4.20 Multi-Agent: 64.0 (#59)
| Benchmark | Gemini 2.5 Pro | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Text | 1458 | 1450 |
| LMArena Creative Writing | 1454 | 1436 |
| LMArena Multi-Turn | 1453 | 1452 |
| Short-Story Creative Writing | 83.8% | — |
| EQ-Bench Creative Writing | 1421 | — |
| WildBench | 85.7% | — |
| LiveBench Language | 67.8% | — |
Frequently asked questions
Is Gemini 2.5 Pro better than Grok 4.20 Multi-Agent?
Grok 4.20 Multi-Agent is the stronger model overall, scoring 46.2 to 45.0 on the Noometry Index.
Which is cheaper, Gemini 2.5 Pro or Grok 4.20 Multi-Agent?
Grok 4.20 Multi-Agent is cheaper. It lists at $1.25 per million input tokens and $2.50 per million output tokens; Gemini 2.5 Pro lists at $1.25 and $10.
Is Gemini 2.5 Pro or Grok 4.20 Multi-Agent better for coding?
They score almost the same on coding (42.4 vs 43.0); test both on your own repository before choosing.
Which has the bigger context window?
Gemini 2.5 Pro does, with 1.05M tokens against 1M.
How many benchmarks do Gemini 2.5 Pro and Grok 4.20 Multi-Agent share?
19 benchmarks have published results for both models. Gemini 2.5 Pro has 78 scored results on Noometry and Grok 4.20 Multi-Agent has 20.