Model comparison
Grok 3 vs Sonar
Grok 3 is the stronger model overall, scoring 39.9 to 38.5 on the Noometry Index.
Last verified . 0 shared benchmarks.
PerplexitySummary
- The widest gap is in reasoning, where Sonar leads 21.1 to 13.7.
Side by side
| Grok 3 | Sonar | |
|---|---|---|
| Provider | xAI | Perplexity |
| Noometry Index | 39.9 | 38.5 |
| Released | 2025-04-09 | 2024-01-01 |
| Weights | Proprietary | Proprietary |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $1 |
| Output $ / M tokens | — | $1 |
| Results tracked | 40 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 3 leads
Grok 3: 41.9 (#115), Sonar: 35.7 (#221)
| Benchmark | Grok 3 | Sonar |
|---|---|---|
| Aider Polyglot | 53.3% | — |
| WeirdML | 37.2% | — |
| LiveBench Coding | — | 35.1% |
| LMArena Coding | 1432 | — |
Agentic & Tool Use Not comparable
Grok 3: 30.5 (#76), Sonar: —
| Benchmark | Grok 3 | Sonar |
|---|---|---|
| BALROG | 29.5% | — |
Reasoning Sonar leads
Grok 3: 13.7 (#333), Sonar: 21.1 (#227)
| Benchmark | Grok 3 | Sonar |
|---|---|---|
| ARC-AGI-2 | 0% | — |
| SimpleBench | 36.1% | — |
| Kagi LLM Benchmark | 61.3% | — |
| ARC-AGI-1 | 5.5% | — |
| LiveBench Reasoning | — | 46.3% |
| LMArena Hard Prompts | 1434 | — |
| LiveBench Data Analysis | — | 37.9% |
| Epoch Capabilities Index | 138.33 | — |
| LiveBench | — | 46.9% |
Math Grok 3 leads
Grok 3: 38.0 (#145), Sonar: 33.7 (#200)
| Benchmark | Grok 3 | Sonar |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 55.6% | — |
| Omni-MATH | 46.4% | — |
| LiveBench Math | — | 41.6% |
| LMArena Math | 1391 | — |
| MATH Level 5 | 88.7% | — |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Not comparable
Grok 3: 46.2 (#82), Sonar: —
| Benchmark | Grok 3 | Sonar |
|---|---|---|
| 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
Grok 3: 52.3 (#87), Sonar: —
| Benchmark | Grok 3 | Sonar |
|---|---|---|
| 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 Grok 3 leads
Grok 3: 75.0 (#73), Sonar: 71.4 (#150)
| Benchmark | Grok 3 | Sonar |
|---|---|---|
| LiveBench Instruction Following | — | 76.2% |
| IFEval | 88.4% | — |
| LMArena Instruction Following | 1409 | — |
Long Context Not comparable
Grok 3: 38.7 (#192), Sonar: —
| Benchmark | Grok 3 | Sonar |
|---|---|---|
| Fiction.LiveBench | 58.3% | — |
| LMArena Longer Query | 1439 | — |
Writing & Preference Grok 3 leads
Grok 3: 55.8 (#141), Sonar: 52.6 (#167)
| Benchmark | Grok 3 | Sonar |
|---|---|---|
| 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 | — |
| LiveBench Language | — | 44.1% |
Frequently asked questions
Is Grok 3 better than Sonar?
Grok 3 is the stronger model overall, scoring 39.9 to 38.5 on the Noometry Index.
Is Grok 3 or Sonar better for coding?
Grok 3 scores higher on coding benchmarks: 41.9 versus 35.7 in the Noometry coding category.
How many benchmarks do Grok 3 and Sonar share?
0 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and Sonar has 7.