Claude Sonnet 4 vs Qwen3 8B
Wins 6 of 6 areas
Coding · Agents · Reasoning · Facts · Long documents · Following instructions
Wins 0 of 6 areas
—
Claude Sonnet 4 is the stronger all-rounder.Qwen3 8B is cheaper.
Scores updated · 22 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 software30Claude Sonnet 43 of 3 tests
- ReasoningHard problems that need careful thinking20Claude Sonnet 42 of 3 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own20Claude Sonnet 42 of 2 tests
- FactsGetting facts right instead of making them up21Claude Sonnet 42 of 3 tests
- Long documentsFinding answers in very long texts10Claude Sonnet 41 of 1 test
- Following instructionsDoing exactly what it is asked10Claude Sonnet 41 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.
Qwen3 8B costs 95% 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 Claude Sonnet 4 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+61.6points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+34.1points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+28.8points ahead
Where Qwen3 8B pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 22 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Sonnet 4
- Terminal-Bench 2.1Claude Sonnet 4 by 34.136.32.2+34.1
- Terminal-Bench HardClaude Sonnet 4 by 28.831.12.3+28.8
- SciCodeClaude Sonnet 4 by 12.14027.9+12.1
AgentsClaude Sonnet 4
- τ-Bench V3 · BankingClaude Sonnet 4 by 1216.74.7+12
- GDPValClaude Sonnet 4 by 9.79.70+9.7
ReasoningClaude Sonnet 4
- GPQA DiamondClaude Sonnet 4 by 18.877.758.9+18.8
- Humanity's Last ExamClaude Sonnet 4 by 6.810.73.9+6.8
- CritPttie0.30tie
FactsClaude Sonnet 4
- AA-Omniscience · Non-hallucinationClaude Sonnet 4 by 61.670.99.3+61.6
- AA-Omniscience · AccuracyClaude Sonnet 4 by 9.122.713.6+9.1
- Vectara HHEM hallucination ratelower is betterQwen3 8B by 5.510.34.8+5.5
Following instructionsClaude Sonnet 4
- IFBenchClaude Sonnet 4 by 21.254.733.5+21.2
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.
- AA-OmniscienceClaude Sonnet 4 by 64.90.2-64.7+64.9
- τ²-Bench Telecom (AA run)Claude Sonnet 4 by 36.864.627.8+36.8
- AIME 2024Qwen3 8B by 32.643.476+32.6
- Artificial Analysis Coding IndexClaude Sonnet 4 by 28.637.69+28.6
- AA IntelligenceClaude Sonnet 4 by 11.618.97.3+11.6
- vectara_avg_summary_lengthClaude Sonnet 4 by 62.2 rating points145.883.6+62.2 rating
- vectara_factual_consistencyQwen3 8B by 5.589.795.2+5.5
- AIME 2025Claude Sonnet 4 by 3.270.567.3+3.2
- vectara_answer_rateQwen3 8B by 1.398.699.9+1.3
Questions people ask
Which is better, Claude Sonnet 4 or Qwen3 8B?
Claude Sonnet 4 wins all six areas where both have results: coding, agents, reasoning, facts, long documents and following instructions. Qwen3 8B wins none, but costs 95% less.
Which is better for coding?
Claude Sonnet 4. It wins 3 of the 3 coding tests both models report; Qwen3 8B wins none.
Which is cheaper?
Claude Sonnet 4 costs $3.00 per million input tokens and $15.00 per million output tokens; Qwen3 8B costs $0.18 and $0.70. That makes Qwen3 8B about 95% cheaper for the same work.
How do you compare the two?
We use the 22 benchmark tests both models have published scores on. The verdict counts the 13 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.