Model comparison
DeepSeek-R1-Distill-Llama-70B vs o3-pro
o3-pro is the stronger model overall, scoring 42.9 to 37.8 on the Noometry Index.
Last verified . 1 shared benchmarks.
DeepSeekSummary
- They share 1 benchmark with published results for both. DeepSeek-R1-Distill-Llama-70B scores higher in 2 categories and o3-pro in 2 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in coding, where o3-pro leads 55.5 to 36.8.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 52.3% for DeepSeek-R1-Distill-Llama-70B and 72.1% for o3-pro.
- DeepSeek-R1-Distill-Llama-70B has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-R1-Distill-Llama-70B | o3-pro | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 37.8 | 42.9 |
| Released | 2025-01-20 | 2025-06-10 |
| Weights | Open | Proprietary |
| Context window | — | 200K |
| Max output | — | 100K |
| Input $ / M tokens | — | $20 |
| Output $ / M tokens | — | $80 |
| Results tracked | 13 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3-pro leads
DeepSeek-R1-Distill-Llama-70B: 36.8 (#202), o3-pro: 55.5 (#24)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | o3-pro |
|---|---|---|
| Aider Polyglot | — | 84.9% |
| WeirdML | — | 58.2% |
| BigCodeBench Instruct | 35.3% | — |
| LiveBench Coding | 51.6% | — |
| BigCodeBench Complete | 49.9% | — |
Reasoning DeepSeek-R1-Distill-Llama-70B leads
DeepSeek-R1-Distill-Llama-70B: 24.9 (#156), o3-pro: 23.8 (#171)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | o3-pro |
|---|---|---|
| Kagi LLM Benchmark | 52.3% | 72.1% |
| ARC-AGI-2 | — | 4.9% |
| ARC-AGI-1 | — | 59.3% |
| LiveBench Reasoning | 67.6% | — |
| DTBench | — | 86.9% |
| LiveBench Data Analysis | 55.9% | — |
| LMCA | — | 38.5% |
| Epoch Capabilities Index | — | 147.42 |
| LiveBench | 54.5% | — |
Math Not comparable
DeepSeek-R1-Distill-Llama-70B: 36.0 (#176), o3-pro: —
| Benchmark | DeepSeek-R1-Distill-Llama-70B | o3-pro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 51.4% | — |
| LiveBench Math | 58.1% | — |
| MATH Level 5 | 89.9% | — |
Knowledge DeepSeek-R1-Distill-Llama-70B leads
DeepSeek-R1-Distill-Llama-70B: 30.7 (#225), o3-pro: 29.5 (#238)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | o3-pro |
|---|---|---|
| GPQA Diamond | 55.7% | — |
| Confabulations | — | 14.2% |
| Vectara Hallucination Rate | — | 23.3% |
Instruction Following Not comparable
DeepSeek-R1-Distill-Llama-70B: 68.2 (#190), o3-pro: —
| Benchmark | DeepSeek-R1-Distill-Llama-70B | o3-pro |
|---|---|---|
| LiveBench Instruction Following | 69.9% | — |
Long Context Not comparable
DeepSeek-R1-Distill-Llama-70B: —, o3-pro: 72.2 (#1)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | o3-pro |
|---|---|---|
| Fiction.LiveBench | — | 97.2% |
Writing & Preference o3-pro leads
DeepSeek-R1-Distill-Llama-70B: 49.0 (#194), o3-pro: 57.1 (#133)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | o3-pro |
|---|---|---|
| Short-Story Creative Writing | — | 84.4% |
| LiveBench Language | 23.8% | — |
Frequently asked questions
Is DeepSeek-R1-Distill-Llama-70B better than o3-pro?
o3-pro is the stronger model overall, scoring 42.9 to 37.8 on the Noometry Index.
Is DeepSeek-R1-Distill-Llama-70B or o3-pro better for coding?
o3-pro scores higher on coding benchmarks: 55.5 versus 36.8 in the Noometry coding category.
How many benchmarks do DeepSeek-R1-Distill-Llama-70B and o3-pro share?
1 benchmark has published results for both models. DeepSeek-R1-Distill-Llama-70B has 13 scored results on Noometry and o3-pro has 12.