Model comparison
DeepSeek-R1-Distill-Llama-70B vs Sonar
DeepSeek-R1-Distill-Llama-70B and Sonar score almost the same on the Noometry Index (37.8 vs 38.5), so choose on price, context window or the category you care about most.
Last verified . 7 shared benchmarks.
DeepSeek
PerplexitySummary
- They share 7 benchmarks with published results for both. DeepSeek-R1-Distill-Llama-70B scores higher in 3 categories and Sonar in 2 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek-R1-Distill-Llama-70B leads 24.9 to 21.1.
- The biggest single-benchmark swing is LiveBench Reasoning: 67.6% for DeepSeek-R1-Distill-Llama-70B and 46.3% for Sonar.
- DeepSeek-R1-Distill-Llama-70B has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-R1-Distill-Llama-70B | Sonar | |
|---|---|---|
| Provider | DeepSeek | Perplexity |
| Noometry Index | 37.8 | 38.5 |
| Released | 2025-01-20 | 2024-01-01 |
| Weights | Open | Proprietary |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $1 |
| Output $ / M tokens | — | $1 |
| Results tracked | 13 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1-Distill-Llama-70B leads
DeepSeek-R1-Distill-Llama-70B: 36.8 (#202), Sonar: 35.7 (#221)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Sonar |
|---|---|---|
| LiveBench Coding | 51.6% | 35.1% |
| BigCodeBench Instruct | 35.3% | — |
| BigCodeBench Complete | 49.9% | — |
Reasoning DeepSeek-R1-Distill-Llama-70B leads
DeepSeek-R1-Distill-Llama-70B: 24.9 (#156), Sonar: 21.1 (#227)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Sonar |
|---|---|---|
| LiveBench Reasoning | 67.6% | 46.3% |
| LiveBench Data Analysis | 55.9% | 37.9% |
| LiveBench | 54.5% | 46.9% |
| Kagi LLM Benchmark | 52.3% | — |
Math DeepSeek-R1-Distill-Llama-70B leads
DeepSeek-R1-Distill-Llama-70B: 36.0 (#176), Sonar: 33.7 (#200)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Sonar |
|---|---|---|
| LiveBench Math | 58.1% | 41.6% |
| OTIS Mock AIME 2024-2025 | 51.4% | — |
| MATH Level 5 | 89.9% | — |
Knowledge Not comparable
DeepSeek-R1-Distill-Llama-70B: 30.7 (#225), Sonar: —
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Sonar |
|---|---|---|
| GPQA Diamond | 55.7% | — |
Instruction Following Sonar leads
DeepSeek-R1-Distill-Llama-70B: 68.2 (#190), Sonar: 71.4 (#150)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Sonar |
|---|---|---|
| LiveBench Instruction Following | 69.9% | 76.2% |
Writing & Preference Sonar leads
DeepSeek-R1-Distill-Llama-70B: 49.0 (#194), Sonar: 52.6 (#167)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Sonar |
|---|---|---|
| LiveBench Language | 23.8% | 44.1% |
Frequently asked questions
Is DeepSeek-R1-Distill-Llama-70B better than Sonar?
DeepSeek-R1-Distill-Llama-70B and Sonar score almost the same on the Noometry Index (37.8 vs 38.5), so choose on price, context window or the category you care about most.
Is DeepSeek-R1-Distill-Llama-70B or Sonar better for coding?
DeepSeek-R1-Distill-Llama-70B scores higher on coding benchmarks: 36.8 versus 35.7 in the Noometry coding category.
How many benchmarks do DeepSeek-R1-Distill-Llama-70B and Sonar share?
7 benchmarks have published results for both models. DeepSeek-R1-Distill-Llama-70B has 13 scored results on Noometry and Sonar has 7.