Scores updated · 13 tests both models report · How we compare
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
DeepSeek-R1-Distill-Qwen-7B$0.15 to read · $0.15 to writePrice from nscale
Granite 3.1 2B InstructNo current price is tracked.
Every test, side by side
All 13 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Other results13 tests
- MATH-500 (EM)DeepSeek-R1-Distill-Qwen-7B by 57.792.835.1+57.7
- AttaQGranite 3.1 2B Instruct by 42.242.584.7+42.2
- DROPDeepSeek-R1-Distill-Qwen-7B by 30.851.821+30.8
- Arena HardGranite 3.1 2B Instruct by 12.910.423.3+12.9
- TruthfulQAGranite 3.1 2B Instruct by 12.747.159.8+12.7
- AlpacaEval-2.0Granite 3.1 2B Instruct by 11.815.327.2+11.8
- GSM8KDeepSeek-R1-Distill-Qwen-7B by 10.978.567.5+10.9
- PopQAGranite 3.1 2B Instruct by 10.79.920.6+10.7
- MMLUGranite 3.1 2B Instruct by 6.450.757.1+6.4
- BigBenchHardDeepSeek-R1-Distill-Qwen-7B by 5.667.461.8+5.6
- HumanEval+DeepSeek-R1-Distill-Qwen-7B by 3.178.475.3+3.1
- IFEvalGranite 3.1 2B Instruct by 3.160.563.6+3.1
- HumanEvaltie79.979.5tie
Questions people ask
How do you compare the two?
We use the 13 benchmark tests both models have published scores on. None of them is in the eight capability areas we count, so this page lists them without a verdict. Each score is the one shown on the model's own page.