Gemma 3 12B vs Qwen3 8B
Wins 2 of 6 areas
Long documents · Following instructions
Wins 4 of 6 areas
Coding · Agents · Reasoning · Facts
Qwen3 8B is the stronger all-rounder.Gemma 3 12B is cheaper and better at long documents.
Scores updated · 23 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 software03Qwen3 8B3 of 3 tests
- FactsGetting facts right instead of making them up02Qwen3 8B2 of 3 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own01Qwen3 8B1 of 3 tests · 2 ties
- ReasoningHard problems that need careful thinking01Qwen3 8B1 of 3 tests · 2 ties
- Long documentsFinding answers in very long texts10Gemma 3 12B1 of 1 test
- Following instructionsDoing exactly what it is asked10Gemma 3 12B1 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.
Gemma 3 12B costs 77% 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 Qwen3 8B pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+24points ahead
- Code for real scientific research problemsSciCode+11.5points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+6.7points ahead
Where Gemma 3 12B pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+3.2points ahead
Every test, side by side
All 23 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingQwen3 8B
- SciCodeQwen3 8B by 11.516.427.9+11.5
- Terminal-Bench 2.1Qwen3 8B by 2.202.2+2.2
- Terminal-Bench HardQwen3 8B by 1.50.82.3+1.5
AgentsQwen3 8B
- τ-Bench V3 · BankingQwen3 8B by 3.90.84.7+3.9
- GDPValtie00tie
- Terminal-Bench 4.0tie00tie
ReasoningQwen3 8B
- GPQA DiamondQwen3 8B by 2434.958.9+24
- Humanity's Last Examtie4.23.9tie
- CritPttie00tie
FactsQwen3 8B
- AA-Omniscience · Non-hallucinationQwen3 8B by 6.72.69.3+6.7
- AA-Omniscience · AccuracyQwen3 8B by 3.310.313.6+3.3
- Vectara HHEM hallucination ratelower is bettertie4.44.8tie
Following instructionsGemma 3 12B
- IFBenchGemma 3 12B by 3.236.733.5+3.2
Other results9 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.
- τ²-Bench Telecom (AA run)Qwen3 8B by 1710.827.8+17
- AA-OmniscienceQwen3 8B by 12.4-77.1-64.7+12.4
- vectara_avg_summary_lengthGemma 3 12B by 6.189.783.6+6.1
- MBPPQwen3 8B by 47377+4
- AA IntelligenceQwen3 8B by 3.53.87.3+3.5
- Artificial Analysis Coding IndexQwen3 8B by 3.25.89+3.2
- vectara_answer_rateQwen3 8B by 2.597.499.9+2.5
- HumanEvaltie85.484.8tie
- vectara_factual_consistencytie95.695.2tie
Questions people ask
Which is better, Gemma 3 12B or Qwen3 8B?
Qwen3 8B wins four of the six areas where both have results: coding, agents, reasoning and facts. Gemma 3 12B wins long documents and following instructions, and costs 77% less.
Which is better for coding?
Qwen3 8B. It wins 3 of the 3 coding tests both models report; Gemma 3 12B wins none.
Which is cheaper?
Gemma 3 12B costs $0.05 per million input tokens and $0.15 per million output tokens; Qwen3 8B costs $0.18 and $0.70. That makes Gemma 3 12B about 77% cheaper for the same work.
How do you compare the two?
We use the 23 benchmark tests both models have published scores on. The verdict counts the 14 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 9 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.