Qwen3 1.7B vs Qwen3 4B 2507 Instruct
Wins 0 of 6 areas
—
Wins 5 of 6 areas
Coding · Reasoning · Facts · Long documents · Following instructions
Qwen3 4B 2507 Instruct is the stronger all-rounder.
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 software03Qwen3 4B 2507 Instruct3 of 3 tests
- FactsGetting facts right instead of making them up02Qwen3 4B 2507 Instruct2 of 2 tests
- ReasoningHard problems that need careful thinking01Qwen3 4B 2507 Instruct1 of 3 tests · 2 ties
- Long documentsFinding answers in very long texts01Qwen3 4B 2507 Instruct1 of 1 test
- Following instructionsDoing exactly what it is asked01Qwen3 4B 2507 Instruct1 of 1 test
- AgentsCarrying out multi-step tasks on its own00Even0 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.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Qwen3 4B 2507 Instruct pulls ahead
- Recent programming contest problemsLiveCodeBench v6+24.6points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+16.1points ahead
- Code for real scientific research problemsSciCode+13.8points ahead
Where Qwen3 1.7B pulls ahead
No clear win on a test scored out of 100.
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.
CodingQwen3 4B 2507 Instruct
- LiveCodeBench v6Qwen3 4B 2507 Instruct by 24.624.148.7+24.6
- SciCodeQwen3 4B 2507 Instruct by 13.84.318.1+13.8
- Terminal-Bench HardQwen3 4B 2507 Instruct by 4.504.5+4.5
ReasoningQwen3 4B 2507 Instruct
- GPQA DiamondQwen3 4B 2507 Instruct by 16.135.651.7+16.1
- Humanity's Last Examtie4.64.5tie
- CritPttie00tie
FactsQwen3 4B 2507 Instruct
- AA-Omniscience · Non-hallucinationQwen3 4B 2507 Instruct by 13.15.218.3+13.1
- AA-Omniscience · AccuracyQwen3 4B 2507 Instruct by 28.910.9+2
Long documentsQwen3 4B 2507 Instruct
- AA-LCRQwen3 4B 2507 Instruct by 11.3011.3+11.3
Following instructionsQwen3 4B 2507 Instruct
- IFBenchQwen3 4B 2507 Instruct by 6.626.933.5+6.6
Other results13 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.
- LCB v5Qwen3 4B 2507 Instruct by 24.326.550.8+24.3
- HumanEval+Qwen3 4B 2507 Instruct by 21.36182.3+21.3
- GSM8KQwen3 4B 2507 Instruct by 17.151.468.5+17.1
- AA-OmniscienceQwen3 4B 2507 Instruct by 15.6-77.5-61.9+15.6
- MGSMQwen3 4B 2507 Instruct by 15.266.681.8+15.2
- MMMLUQwen3 4B 2507 Instruct by 14.246.560.7+14.2
- MMLUQwen3 4B 2507 Instruct by 13.259.172.3+13.2
- IFEvalQwen3 4B 2507 Instruct by 11.67485.6+11.6
- MMLU-ProQwen3 1.7B by 4.75752.3+4.7
- MATH-500 (EM)Qwen3 4B 2507 Instruct by 3.781.985.6+3.7
- AA IntelligenceQwen3 4B 2507 Instruct by 1.55.26.7+1.5
- τ²-Bench Telecom (AA run)tie2626.6tie
- AA Agentic Indextie8.78.9tie
Questions people ask
Which is better, Qwen3 1.7B or Qwen3 4B 2507 Instruct?
Qwen3 4B 2507 Instruct wins five of the six areas where both have results: coding, reasoning, facts, long documents and following instructions. Qwen3 1.7B wins none. They are level on agents.
Which is better for coding?
Qwen3 4B 2507 Instruct. It wins 3 of the 3 coding tests both models report; Qwen3 1.7B wins none.
How do you compare the two?
We use the 24 benchmark tests both models have published scores on. The verdict counts the 11 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 13 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.