Llama 3.1 Nemotron Instruct 70B vs Llama 3 Instruct 8B
Wins 5 of 5 areas
Coding · Reasoning · Facts · Long documents · Following instructions
Wins 0 of 5 areas
—
Llama 3.1 Nemotron Instruct 70B is the stronger all-rounder.
Scores updated · 19 tests both models report · How we compare
Where each one wins
Tests won in each of the five areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- CodingWriting and fixing software20Llama 3.1 Nemotron Instruct 70B2 of 2 tests
- FactsGetting facts right instead of making them up20Llama 3.1 Nemotron Instruct 70B2 of 2 tests
- ReasoningHard problems that need careful thinking10Llama 3.1 Nemotron Instruct 70B1 of 3 tests · 2 ties
- Long documentsFinding answers in very long texts10Llama 3.1 Nemotron Instruct 70B1 of 1 test
- Following instructionsDoing exactly what it is asked10Llama 3.1 Nemotron Instruct 70B1 of 1 test
Long documents and following instructions rest on a single test each.
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.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Llama 3.1 Nemotron Instruct 70B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+18.5points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+16.9points ahead
- Code for real scientific research problemsSciCode+11.4points ahead
Where Llama 3 Instruct 8B pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 19 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingLlama 3.1 Nemotron Instruct 70B
- SciCodeLlama 3.1 Nemotron Instruct 70B by 11.423.311.9+11.4
- Terminal-Bench HardLlama 3.1 Nemotron Instruct 70B by 4.54.50+4.5
ReasoningLlama 3.1 Nemotron Instruct 70B
- GPQA DiamondLlama 3.1 Nemotron Instruct 70B by 16.946.529.6+16.9
- Humanity's Last Examtie4.25.1tie
- CritPttie00tie
FactsLlama 3.1 Nemotron Instruct 70B
- AA-Omniscience · Non-hallucinationLlama 3.1 Nemotron Instruct 70B by 18.528.710.2+18.5
- AA-Omniscience · AccuracyLlama 3.1 Nemotron Instruct 70B by 7.417.810.4+7.4
Long documentsLlama 3.1 Nemotron Instruct 70B
- AA-LCRLlama 3.1 Nemotron Instruct 70B by 8.38.30+8.3
Following instructionsLlama 3.1 Nemotron Instruct 70B
- IFBenchLlama 3.1 Nemotron Instruct 70B by 6.130.724.6+6.1
Other results10 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- GSM8KLlama 3.1 Nemotron Instruct 70B by 41.591.449.9+41.5
- AA-OmniscienceLlama 3.1 Nemotron Instruct 70B by 29.3-40.8-70.1+29.3
- τ²-Bench Telecom (AA run)Llama 3.1 Nemotron Instruct 70B by 23.123.10+23.1
- WinoGrandeLlama 3.1 Nemotron Instruct 70B by 19.584.565+19.5
- HellaSwagLlama 3.1 Nemotron Instruct 70B by 14.585.671.1+14.5
- ARC-ChallengeLlama 3 Instruct 8B by 13.669.282.8+13.6
- AA Agentic IndexLlama 3.1 Nemotron Instruct 70B by 7.77.70+7.7
- Artificial Analysis Coding IndexLlama 3.1 Nemotron Instruct 70B by 6.810.84+6.8
- TruthfulQALlama 3 Instruct 8B by 4.658.663.2+4.6
- AA IntelligenceLlama 3.1 Nemotron Instruct 70B by 2.16.94.8+2.1
Questions people ask
Which is better, Llama 3.1 Nemotron Instruct 70B or Llama 3 Instruct 8B?
Llama 3.1 Nemotron Instruct 70B wins all five areas where both have results: coding, reasoning, facts, long documents and following instructions. Llama 3 Instruct 8B wins none.
Which is better for coding?
Llama 3.1 Nemotron Instruct 70B. It wins 2 of the 2 coding tests both models report; Llama 3 Instruct 8B wins none.
How do you compare the two?
We use the 19 benchmark tests both models have published scores on. The verdict counts the 9 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 10 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.