Claude 3 Haiku vs Qwen3 Max (Non-Reasoning)
Wins 0 of 6 areas
—
Wins 5 of 6 areas
Coding · Agents · Reasoning · Long documents · Following instructions
Qwen3 Max (Non-Reasoning) is the stronger all-rounder.Claude 3 Haiku is cheaper.
Scores updated · 15 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.
- ReasoningHard problems that need careful thinking02Qwen3 Max (Non-Reasoning)2 of 3 tests · 1 tie
- CodingWriting and fixing software02Qwen3 Max (Non-Reasoning)2 of 2 tests
- AgentsCarrying out multi-step tasks on its own01Qwen3 Max (Non-Reasoning)1 of 1 test
- Long documentsFinding answers in very long texts01Qwen3 Max (Non-Reasoning)1 of 1 test
- Following instructionsDoing exactly what it is asked01Qwen3 Max (Non-Reasoning)1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
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.
Claude 3 Haiku costs 79% 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 Max (Non-Reasoning) pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+39points ahead
- Reasons across sets of long documentsAA-LCR+22.3points ahead
- Code for real scientific research problemsSciCode+19.7points ahead
Where Claude 3 Haiku pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+9.4points ahead
Every test, side by side
All 15 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingQwen3 Max (Non-Reasoning)
- SciCodeQwen3 Max (Non-Reasoning) by 19.718.638.3+19.7
- Terminal-Bench HardQwen3 Max (Non-Reasoning) by 19.70.820.5+19.7
ReasoningQwen3 Max (Non-Reasoning)
- GPQA DiamondQwen3 Max (Non-Reasoning) by 3937.476.4+39
- Humanity's Last ExamQwen3 Max (Non-Reasoning) by 7.84.111.9+7.8
- CritPttie00tie
FactsEven
- AA-Omniscience · Non-hallucinationClaude 3 Haiku by 9.419.510.1+9.4
- AA-Omniscience · AccuracyQwen3 Max (Non-Reasoning) by 6.817.624.4+6.8
Long documentsQwen3 Max (Non-Reasoning)
- AA-LCRQwen3 Max (Non-Reasoning) by 22.327.750+22.3
Following instructionsQwen3 Max (Non-Reasoning)
- IFBenchQwen3 Max (Non-Reasoning) by 836.144.1+8
Other results5 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 Max (Non-Reasoning) by 53.221.174.3+53.2
- AA Agentic IndexQwen3 Max (Non-Reasoning) by 36743+36
- Artificial Analysis Coding IndexQwen3 Max (Non-Reasoning) by 19.76.726.4+19.7
- AA IntelligenceQwen3 Max (Non-Reasoning) by 105.615.6+10
- AA-OmniscienceQwen3 Max (Non-Reasoning) by 5.1-48.6-43.5+5.1
Questions people ask
Which is better, Claude 3 Haiku or Qwen3 Max (Non-Reasoning)?
Qwen3 Max (Non-Reasoning) wins five of the six areas where both have results: coding, agents, reasoning, long documents and following instructions. Claude 3 Haiku wins none, but costs 79% less. They are level on facts.
Which is better for coding?
Qwen3 Max (Non-Reasoning). It wins 2 of the 2 coding tests both models report; Claude 3 Haiku wins none.
Which is cheaper?
Claude 3 Haiku costs $0.25 per million input tokens and $1.25 per million output tokens; Qwen3 Max (Non-Reasoning) costs $1.20 and $6.00. That makes Claude 3 Haiku about 79% cheaper for the same work.
How do you compare the two?
We use the 15 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 5 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.