Llama 3.1 8B Instruct vs Nova Micro
Wins 2 of 6 areas
Coding · Long documents
Wins 1 of 6 areas
Reasoning
Llama 3.1 8B Instruct wins more areas, narrowly.Nova Micro is better at reasoning.
Scores updated · 24 tests both models report · How we compare
Where each one wins
Tests won in each of the six 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 software10Llama 3.1 8B Instruct1 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts10Llama 3.1 8B Instruct1 of 1 test
- ReasoningHard problems that need careful thinking01Nova Micro1 of 3 tests · 2 ties
- AgentsCarrying out multi-step tasks on its own00Even0 each · 1 tie
- FactsGetting facts right instead of making them up11Even1 each
- Following instructionsDoing exactly what it is asked00Even0 each · 1 tie
Agents, 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.
Llama 3.1 8B Instruct costs 60% less for the same work.
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 8B Instruct pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+22.1points ahead
- Reasons across sets of long documentsAA-LCR+5points ahead
- Code for real scientific research problemsSciCode+3.8points ahead
Where Nova Micro pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+9.9points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+1.5points ahead
Every test, side by side
All 24 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingLlama 3.1 8B Instruct
- SciCodeLlama 3.1 8B Instruct by 3.813.29.4+3.8
- Terminal-Bench Hardtie0.81.5tie
ReasoningNova Micro
- GPQA DiamondNova Micro by 9.925.935.8+9.9
- Humanity's Last Examtie5.34.6tie
- CritPttie00tie
FactsEven
- AA-Omniscience · Non-hallucinationLlama 3.1 8B Instruct by 22.15734.9+22.1
- AA-Omniscience · AccuracyNova Micro by 1.58.510+1.5
Long documentsLlama 3.1 8B Instruct
- AA-LCRLlama 3.1 8B Instruct by 51813+5
Other results14 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.
- BFCLv4Llama 3.1 8B Instruct by 19.976.156.2+19.9
- DROPNova Micro by 19.859.579.3+19.8
- AA-OmniscienceLlama 3.1 8B Instruct by 17.6-30.9-48.5+17.6
- OpenBookQANova Micro by 14.87488.8+14.8
- GSM8KNova Micro by 12.579.892.3+12.5
- MMLULlama 3.1 8B Instruct by 97364+9
- HumanEvalNova Micro by 8.572.681.1+8.5
- naturalquestions_closedbookNova Micro by 7.620.928.5+7.6
- ARC-ChallengeNova Micro by 6.883.490.2+6.8
- IFEvalNova Micro by 6.880.487.2+6.8
- τ²-Bench Telecom (AA run)Llama 3.1 8B Instruct by 2.416.414+2.4
- Artificial Analysis Coding IndexLlama 3.1 8B Instruct by 1.35.44.1+1.3
- NarrativeQALlama 3.1 8B Instruct by 1.275.674.4+1.2
- AA IntelligenceLlama 3.1 8B Instruct by 16.95.9+1
Questions people ask
Which is better, Llama 3.1 8B Instruct or Nova Micro?
Llama 3.1 8B Instruct wins two of the six areas where both have results: coding and long documents. Nova Micro wins reasoning. They are level on agents, facts and following instructions.
Which is better for coding?
Llama 3.1 8B Instruct. It wins 1 of the 2 coding tests both models report; Nova Micro wins none, and 1 is a tie.
Which is cheaper?
Llama 3.1 8B Instruct costs $0.02 per million input tokens and $0.05 per million output tokens; Nova Micro costs $0.04 and $0.14. That makes Llama 3.1 8B Instruct about 60% cheaper for the same work.
How do you compare the two?
We use the 24 benchmark tests both models have published scores on. The verdict counts the 10 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 14 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.