Claude 3.5 Haiku vs Qwen3 Coder 30B A3B Instruct
Wins 1 of 6 areas
Following instructions
Wins 4 of 6 areas
Coding · Agents · Reasoning · Long documents
Qwen3 Coder 30B A3B Instruct is the stronger all-rounder.Claude 3.5 Haiku is better at following instructions.
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 thinking01Qwen3 Coder 30B A3B Instruct1 of 3 tests · 2 ties
- CodingWriting and fixing software01Qwen3 Coder 30B A3B Instruct1 of 2 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own01Qwen3 Coder 30B A3B Instruct1 of 1 test
- Long documentsFinding answers in very long texts01Qwen3 Coder 30B A3B Instruct1 of 1 test
- Following instructionsDoing exactly what it is asked10Claude 3.5 Haiku1 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.
Qwen3 Coder 30B A3B Instruct costs 44% 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 Coder 30B A3B Instruct pulls ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+12.9points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+10.8points ahead
- Reasons across sets of long documentsAA-LCR+5.4points ahead
Where Claude 3.5 Haiku pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+39.2points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+10.1points 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 Coder 30B A3B Instruct
- Terminal-Bench HardQwen3 Coder 30B A3B Instruct by 12.92.315.2+12.9
- SciCodetie27.427.8tie
AgentsQwen3 Coder 30B A3B Instruct
- GDPValQwen3 Coder 30B A3B Instruct by 10.5010.5+10.5
ReasoningQwen3 Coder 30B A3B Instruct
- GPQA DiamondQwen3 Coder 30B A3B Instruct by 10.840.851.6+10.8
- Humanity's Last Examtie3.63.8tie
- CritPttie00tie
FactsEven
- AA-Omniscience · Non-hallucinationClaude 3.5 Haiku by 39.258.919.7+39.2
- AA-Omniscience · AccuracyQwen3 Coder 30B A3B Instruct by 3.213.216.4+3.2
Long documentsQwen3 Coder 30B A3B Instruct
- AA-LCRQwen3 Coder 30B A3B Instruct by 5.427.332.7+5.4
Following instructionsClaude 3.5 Haiku
- IFBenchClaude 3.5 Haiku by 10.142.832.7+10.1
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.
- AA-OmniscienceClaude 3.5 Haiku by 28.2-22.5-50.7+28.2
- AA Agentic IndexQwen3 Coder 30B A3B Instruct by 19.21.921.1+19.2
- τ²-Bench Telecom (AA run)Qwen3 Coder 30B A3B Instruct by 9.924.634.5+9.9
- Artificial Analysis Coding IndexQwen3 Coder 30B A3B Instruct by 3.515.919.4+3.5
- AA Intelligencetie8.99.6tie
Questions people ask
Which is better, Claude 3.5 Haiku or Qwen3 Coder 30B A3B Instruct?
Qwen3 Coder 30B A3B Instruct wins four of the six areas where both have results: coding, agents, reasoning and long documents. Claude 3.5 Haiku wins following instructions. They are level on facts.
Which is better for coding?
Qwen3 Coder 30B A3B Instruct. It wins 1 of the 2 coding tests both models report; Claude 3.5 Haiku wins none, and 1 is a tie.
Which is cheaper?
Claude 3.5 Haiku costs $0.80 per million input tokens and $4.00 per million output tokens; Qwen3 Coder 30B A3B Instruct costs $0.45 and $2.25. That makes Qwen3 Coder 30B A3B Instruct about 44% 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.