Model comparison
DeepSeek-R1-Distill-Llama-70B vs DeepSeek-R1-Distill-Qwen-32B
DeepSeek-R1-Distill-Llama-70B is the stronger model overall, scoring 37.8 to 35.5 on the Noometry Index.
Last verified . 11 shared benchmarks.
DeepSeekSummary
- They share 11 benchmarks with published results for both. DeepSeek-R1-Distill-Llama-70B scores higher in 4 categories and DeepSeek-R1-Distill-Qwen-32B in 2 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek-R1-Distill-Llama-70B leads 24.9 to 18.2.
- The biggest single-benchmark swing is LiveBench Coding: 51.6% for DeepSeek-R1-Distill-Llama-70B and 33.7% for DeepSeek-R1-Distill-Qwen-32B.
Side by side
| DeepSeek-R1-Distill-Llama-70B | DeepSeek-R1-Distill-Qwen-32B | |
|---|---|---|
| Provider | DeepSeek | DeepSeek |
| Noometry Index | 37.8 | 35.5 |
| Released | 2025-01-20 | 2025-01-20 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 13 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
DeepSeek-R1-Distill-Llama-70B: 36.8 (#202), DeepSeek-R1-Distill-Qwen-32B: 36.1 (#212)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| BigCodeBench Instruct | 35.3% | 43.9% |
| LiveBench Coding | 51.6% | 33.7% |
| BigCodeBench Complete | 49.9% | 54.9% |
Agentic & Tool Use Not comparable
DeepSeek-R1-Distill-Llama-70B: —, DeepSeek-R1-Distill-Qwen-32B: 28.1 (#94)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| BALROG | — | 19.5% |
Reasoning DeepSeek-R1-Distill-Llama-70B leads
DeepSeek-R1-Distill-Llama-70B: 24.9 (#156), DeepSeek-R1-Distill-Qwen-32B: 18.2 (#284)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| LiveBench Reasoning | 67.6% | 52.3% |
| LiveBench Data Analysis | 55.9% | 45.4% |
| LiveBench | 54.5% | 45.5% |
| Kagi LLM Benchmark | 52.3% | — |
| Chess Puzzles | — | 1% |
| Epoch Capabilities Index | — | 137.44 |
Math DeepSeek-R1-Distill-Llama-70B leads
DeepSeek-R1-Distill-Llama-70B: 36.0 (#176), DeepSeek-R1-Distill-Qwen-32B: 34.5 (#194)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 51.4% | 55.6% |
| LiveBench Math | 58.1% | 59.4% |
| MATH Level 5 | 89.9% | — |
Knowledge DeepSeek-R1-Distill-Qwen-32B leads
DeepSeek-R1-Distill-Llama-70B: 30.7 (#225), DeepSeek-R1-Distill-Qwen-32B: 35.7 (#182)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| GPQA Diamond | 55.7% | 64.1% |
Instruction Following DeepSeek-R1-Distill-Llama-70B leads
DeepSeek-R1-Distill-Llama-70B: 68.2 (#190), DeepSeek-R1-Distill-Qwen-32B: 61.6 (#243)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| LiveBench Instruction Following | 69.9% | 55.7% |
Writing & Preference Too close to call
DeepSeek-R1-Distill-Llama-70B: 49.0 (#194), DeepSeek-R1-Distill-Qwen-32B: 49.6 (#188)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| LiveBench Language | 23.8% | 26.8% |
Frequently asked questions
Is DeepSeek-R1-Distill-Llama-70B better than DeepSeek-R1-Distill-Qwen-32B?
DeepSeek-R1-Distill-Llama-70B is the stronger model overall, scoring 37.8 to 35.5 on the Noometry Index.
Is DeepSeek-R1-Distill-Llama-70B or DeepSeek-R1-Distill-Qwen-32B better for coding?
They score almost the same on coding (36.8 vs 36.1); test both on your own repository before choosing.
How many benchmarks do DeepSeek-R1-Distill-Llama-70B and DeepSeek-R1-Distill-Qwen-32B share?
11 benchmarks have published results for both models. DeepSeek-R1-Distill-Llama-70B has 13 scored results on Noometry and DeepSeek-R1-Distill-Qwen-32B has 14.