Model comparison
DeepSeek-V2.5 (Sep 2024) vs Sonar
DeepSeek-V2.5 (Sep 2024) and Sonar score almost the same on the Noometry Index (37.6 vs 38.5), so choose on price, context window or the category you care about most.
Last verified . 0 shared benchmarks.
DeepSeek
PerplexitySummary
- The widest gap is in reasoning, where DeepSeek-V2.5 (Sep 2024) leads 25.6 to 21.1.
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Sonar | |
|---|---|---|
| Provider | DeepSeek | Perplexity |
| Noometry Index | 37.6 | 38.5 |
| Released | 2024-09-06 | 2024-01-01 |
| Weights | Open | Proprietary |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $1 |
| Output $ / M tokens | — | $1 |
| Results tracked | 22 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Sonar leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Sonar: 35.7 (#221)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Sonar |
|---|---|---|
| Aider Polyglot | 17.8% | — |
| BigCodeBench Instruct | 48.6% | — |
| LiveBench Coding | — | 35.1% |
| LMArena Coding | 1309 | — |
| BigCodeBench Complete | 53.2% | — |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Sonar: 21.1 (#227)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Sonar |
|---|---|---|
| LiveBench Reasoning | — | 46.3% |
| LMArena Hard Prompts | 1289 | — |
| LiveBench Data Analysis | — | 37.9% |
| LiveBench | — | 46.9% |
Math DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Sonar: 33.7 (#200)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Sonar |
|---|---|---|
| LiveBench Math | — | 41.6% |
| LMArena Math | 1288 | — |
Knowledge Not comparable
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Sonar: —
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Sonar |
|---|---|---|
| LMArena Expert | 1266 | — |
Multilingual Not comparable
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Sonar: —
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Sonar |
|---|---|---|
| LMArena Non-English | 1273 | — |
| LMArena Chinese | 1318 | — |
| LMArena French | 1289 | — |
| LMArena German | 1258 | — |
| LMArena Japanese | 1228 | — |
| LMArena Korean | 1209 | — |
| LMArena Russian | 1289 | — |
| LMArena Spanish | 1248 | — |
Instruction Following Sonar leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Sonar: 71.4 (#150)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Sonar |
|---|---|---|
| LiveBench Instruction Following | — | 76.2% |
| LMArena Instruction Following | 1280 | — |
Long Context Not comparable
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Sonar: —
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Sonar |
|---|---|---|
| LMArena Longer Query | 1301 | — |
Writing & Preference Sonar leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Sonar: 52.6 (#167)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Sonar |
|---|---|---|
| LMArena Text | 1294 | — |
| LMArena Creative Writing | 1285 | — |
| LMArena Multi-Turn | 1297 | — |
| LiveBench Language | — | 44.1% |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Sonar?
DeepSeek-V2.5 (Sep 2024) and Sonar score almost the same on the Noometry Index (37.6 vs 38.5), so choose on price, context window or the category you care about most.
Is DeepSeek-V2.5 (Sep 2024) or Sonar better for coding?
Sonar scores higher on coding benchmarks: 35.7 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Sonar share?
0 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Sonar has 7.