# o3 vs Sonar

> o3 is the stronger model overall, scoring 47.4 to 37.8 on the Noometry Index. Sonar costs 3.5× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/o3-vs-sonar
- Last updated: 2026-10-10
- Shared benchmarks: 0

## Summary

- The widest gap is in math, where o3 leads 48.9 to 31.3.
- Sonar is cheaper at $1 / $1 per million input/output tokens, against $2 / $8 for o3.
- o3 accepts more context: 200K tokens versus 128K.

## Snapshot

| | o3 | Sonar |
|---|---|---|
| Provider | OpenAI | Perplexity |
| Noometry Index | 47.4 | 37.8 |
| Rank | 53 | 184 |
| Context | 200K | 128K |
| Input $/M | $2 | $1 |
| Output $/M | $8 | $1 |
| Weights | Proprietary | Proprietary |

## Coding

- o3: 47.3 (#57)
- Sonar: 35.1 (#217)

| Benchmark | o3 | Sonar |
|---|---|---|
| SWE-bench Verified | 62.3% | — |
| SWE-bench Verified (bash only) | 58.4% | — |
| Aider Polyglot | 81.3% | — |
| GSO | 8.8% | — |
| WeirdML | 52.4% | — |
| LiveBench Coding | — | 35.1% |
| LMArena Coding | 1408 | — |
| CadEval | 74% | — |
| ALE-Bench | 933.55 | — |

## Agentic & Tool Use

- o3: 34.3 (#42)
- Sonar: —

| Benchmark | o3 | Sonar |
|---|---|---|
| Berkeley Function Calling Leaderboard | 63% | — |
| GDPval | 30.8% | — |
| DeepResearch Bench | 45.2% | — |
| OSWorld | 23% | — |
| LMArena Search | 1144 | — |
| METR Time Horizons | 65.4% | — |

## Reasoning

- o3: 32.1 (#66)
- Sonar: 20.3 (#221)

| Benchmark | o3 | Sonar |
|---|---|---|
| ARC-AGI-2 | 6.5% | — |
| SimpleBench | 53.1% | — |
| Kagi LLM Benchmark | 67.6% | — |
| ARC-AGI-1 | 60.8% | — |
| CritPt | 1.4% | — |
| Chess Puzzles | 38% | — |
| EnigmaEval | 13.1% | — |
| LiveBench Reasoning | — | 46.3% |
| LMArena Hard Prompts | 1402 | — |
| Mystery Game Puzzles | 29% | — |
| DTBench | 84.8% | — |
| LiveBench Data Analysis | — | 37.9% |
| LMCA | 39.7% | — |
| Epoch Capabilities Index | 146.88 | — |
| ForecastBench | 62.5 | — |
| LiveBench | — | 46.9% |

## Math

- o3: 48.9 (#49)
- Sonar: 31.3 (#197)

| Benchmark | o3 | Sonar |
|---|---|---|
| FrontierMath (Tiers 1-3) | 33.3% | — |
| OTIS Mock AIME 2024-2025 | 84.4% | — |
| Omni-MATH | 71.4% | — |
| LiveBench Math | — | 41.6% |
| LMArena Math | 1426 | — |
| MATH Level 5 | 97.8% | — |
| FrontierMath (Feb 2025 set) | 18.7% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |

## Knowledge

- o3: 54.5 (#47)
- Sonar: —

| Benchmark | o3 | Sonar |
|---|---|---|
| GPQA Diamond | 81.8% | — |
| Humanity's Last Exam | 20.3% | — |
| SimpleQA Verified | 49.4% | — |
| MMLU-Pro | 85.9% | — |
| Confabulations | 14.4% | — |
| GPQA (HELM) | 75.3% | — |
| LMArena Expert | 1403 | — |

## Multimodal

- o3: 40.7 (#35)
- Sonar: —

| Benchmark | o3 | Sonar |
|---|---|---|
| LMArena Vision | 1214 | — |
| GeoBench | 74% | — |
| VPCT | 52% | — |

## Multilingual

- o3: 51.9 (#99)
- Sonar: —

| Benchmark | o3 | Sonar |
|---|---|---|
| LMArena Non-English | 1402 | — |
| LMArena Chinese | 1438 | — |
| LMArena French | 1430 | — |
| LMArena German | 1420 | — |
| LMArena Japanese | 1404 | — |
| LMArena Korean | 1371 | — |
| LMArena Russian | 1407 | — |
| LMArena Spanish | 1395 | — |

## Instruction Following

- o3: 73.0 (#122)
- Sonar: 71.5 (#146)

| Benchmark | o3 | Sonar |
|---|---|---|
| LiveBench Instruction Following | — | 76.2% |
| IFEval | 86.9% | — |
| LMArena Instruction Following | 1368 | — |

## Long Context

- o3: 53.2 (#6)
- Sonar: —

| Benchmark | o3 | Sonar |
|---|---|---|
| Fiction.LiveBench | 88.9% | — |
| CL-bench | 17.8% | — |
| LMArena Longer Query | 1372 | — |

## Writing & Preference

- o3: 63.7 (#62)
- Sonar: 52.6 (#166)

| Benchmark | o3 | Sonar |
|---|---|---|
| LMArena Text | 1410 | — |
| LMArena Creative Writing | 1359 | — |
| Short-Story Creative Writing | 83.9% | — |
| EQ-Bench Creative Writing | 1676 | — |
| WildBench | 86.1% | — |
| LMArena Multi-Turn | 1405 | — |
| LiveBench Language | — | 44.1% |

## FAQ

### Is o3 better than Sonar?

o3 is the stronger model overall, scoring 47.4 to 37.8 on the Noometry Index. Sonar costs 3.5× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.

### Which is cheaper, o3 or Sonar?

Sonar is cheaper. It lists at $1 per million input tokens and $1 per million output tokens; o3 lists at $2 and $8.

### Is o3 or Sonar better for coding?

o3 scores higher on coding benchmarks: 47.3 versus 35.1 in the Noometry coding category.

### Which has the bigger context window?

o3 does, with 200K tokens against 128K.

### How many benchmarks do o3 and Sonar share?

0 benchmarks have published results for both models. o3 has 63 scored results on Noometry and Sonar has 7.