Model comparison
DeepSeek LLM 67B vs Sonar
Sonar is the stronger model overall, scoring 38.5 to 24.9 on the Noometry Index.
Last verified . 0 shared benchmarks.
DeepSeek
PerplexitySummary
- The widest gap is in math, where Sonar leads 33.7 to 8.7.
- DeepSeek LLM 67B has downloadable open weights; the other is API-only.
Side by side
| DeepSeek LLM 67B | Sonar | |
|---|---|---|
| Provider | DeepSeek | Perplexity |
| Noometry Index | 24.9 | 38.5 |
| Released | 2023-11-29 | 2024-01-01 |
| Weights | Open | Proprietary |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $1 |
| Output $ / M tokens | — | $1 |
| Results tracked | 15 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Sonar leads
DeepSeek LLM 67B: 31.9 (#278), Sonar: 35.7 (#221)
| Benchmark | DeepSeek LLM 67B | Sonar |
|---|---|---|
| LiveBench Coding | — | 35.1% |
| LMArena Coding | 1096 | — |
Reasoning Sonar leads
DeepSeek LLM 67B: 16.5 (#304), Sonar: 21.1 (#227)
| Benchmark | DeepSeek LLM 67B | Sonar |
|---|---|---|
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 46.3% |
| LMArena Hard Prompts | 1070 | — |
| LiveBench Data Analysis | — | 37.9% |
| Epoch Capabilities Index | 110.5 | — |
| LiveBench | — | 46.9% |
Math Sonar leads
DeepSeek LLM 67B: 8.7 (#324), Sonar: 33.7 (#200)
| Benchmark | DeepSeek LLM 67B | Sonar |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | — |
| LiveBench Math | — | 41.6% |
| LMArena Math | 1108 | — |
| MATH Level 5 | 6.4% | — |
Knowledge Not comparable
DeepSeek LLM 67B: 7.0 (#313), Sonar: —
| Benchmark | DeepSeek LLM 67B | Sonar |
|---|---|---|
| GPQA Diamond | 24.6% | — |
Multilingual Not comparable
DeepSeek LLM 67B: 29.4 (#267), Sonar: —
| Benchmark | DeepSeek LLM 67B | Sonar |
|---|---|---|
| LMArena Non-English | 1073 | — |
| LMArena Chinese | 1132 | — |
Instruction Following Sonar leads
DeepSeek LLM 67B: 55.4 (#277), Sonar: 71.4 (#150)
| Benchmark | DeepSeek LLM 67B | Sonar |
|---|---|---|
| LiveBench Instruction Following | — | 76.2% |
| LMArena Instruction Following | 1079 | — |
Long Context Not comparable
DeepSeek LLM 67B: 33.1 (#265), Sonar: —
| Benchmark | DeepSeek LLM 67B | Sonar |
|---|---|---|
| LMArena Longer Query | 1092 | — |
Writing & Preference Sonar leads
DeepSeek LLM 67B: 31.6 (#282), Sonar: 52.6 (#167)
| Benchmark | DeepSeek LLM 67B | Sonar |
|---|---|---|
| LMArena Text | 1105 | — |
| LMArena Creative Writing | 1067 | — |
| LMArena Multi-Turn | 1082 | — |
| LiveBench Language | — | 44.1% |
Frequently asked questions
Is DeepSeek LLM 67B better than Sonar?
Sonar is the stronger model overall, scoring 38.5 to 24.9 on the Noometry Index.
Is DeepSeek LLM 67B or Sonar better for coding?
Sonar scores higher on coding benchmarks: 35.7 versus 31.9 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and Sonar share?
0 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Sonar has 7.