Compare / Qwen3-235B-A22B-Instruct-2507 vs Claude Haiku 4.5

Qwen3-235B-A22B-Instruct-2507 vs Claude Haiku 4.5

Which is cheaper, which is more capable, and which does better at each kind of work, from 25 results on 4 sources that measured both models.

AlibabaQwen3-235B-A22B-Instruct-2507 Released 21 Jul 2025 · $0.15 in · $0.75 out
AnthropicClaude Haiku 4.5 Released 15 Oct 2025 · $1.00 in · $5.00 out
Qwen3-235B-A22B-Instruct-2507 wins 75 too close to callClaude Haiku 4.5 wins 13

Of the 25 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.

Results as of 9 October 2026 (catalogue release 2026-10-09-dd620e778ac5). See also Alibaba vs Anthropic, every model from both labs.

Qwen3-235B-A22B-Instruct-2507 or Claude Haiku 4.5? The short answer

Pick Claude Haiku 4.5 if quality matters most: it scores higher (Intelligence Index 16.9 against 12.0). Pick Qwen3-235B-A22B-Instruct-2507 if cost matters more: it costs 6.7 times less per token.

  • Qwen3-235B-A22B-Instruct-2507 is cheaper: $0.30 against $2.00 per million tokens, 6.7 times less.
  • Claude Haiku 4.5 scores higher on the Artificial Analysis Intelligence Index: 16.9 against 12.0.
  • Claude Haiku 4.5 is newer: released 15 Oct 2025, against 21 Jul 2025 for Qwen3-235B-A22B-Instruct-2507.
  • On third-party benchmarks, on the results both models have: Qwen3-235B-A22B-Instruct-2507 does better for reasoning and knowledge and sticking to the facts; Claude Haiku 4.5 for agents and tool use.

Head to head

12.0
Intelligence Index
16.9
$0.30
Price per million tokens, blendedlower is better
$2.00
262,144
Context window, tokens
200,000

Every fact side by side

Qwen3-235B-A22B-Instruct-2507Claude Haiku 4.5
Price per million tokens$0.15 in · $0.75 out$1.00 in · $5.00 out
Blended (3 in : 1 out)$0.30$2.00
Cheapest host, blended$0.15 (GMICloud, FP8)$2.00 (Amazon Bedrock)
Cached input, per million–$0.10 cached input
Intelligence Index12.016.9
Spring Prompt overall–7
Output speed, tokens a second–97
Context window262,144 tokens200,000 tokens
Released21 Jul 202515 Oct 2025 (newer)
WeightsOpenClosed
Developer based inChinathe US

Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.

Which is better for what

Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.

Product listingsTurning a sparse product feed and photos into listings that can go live not measured 19 of 20
Decks from an analysisTurning a finished analysis into a deck you could present as it is not measured 15 of 19
User surveysPlanning a user survey and reading its results without being misled not measured 8 of 8
Marketing planningPlanning a year of ad spend without overspending not measured 18 of 19
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 51 of 164 91 of 164
Reasoning and knowledgeHard questions across science, maths and general knowledge 129 of 198 127 of 198
WritingWhat people prefer in blind comparisons, and judged writing quality 85 of 188 87 of 188
Sticking to the factsSummarising without inventing things, and factual answers 98 of 152 117 of 152
Speed under pressureGood decisions against a real clock (fast chess) not measured 17 of 23

From our research

Every shared result

Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.

MeasureQwen3-235B-A22B-Instruct-2507Claude Haiku 4.5
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Overall · TextArena rating, higher is better 109 · 121 of 387 measured 1,422 1,414
Overall · Text factualityArena rating, higher is better 94 · 118 of 163 measured 1,431 1,418
Business, management and finance · TextArena rating, higher is better 97 · 109 of 380 measured 1,432 1,420
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 153 · 116 of 243 measured 12.0 16.9thinking reasoning
GPQA Diamond% of questions, higher is better 136 · 163 of 231 measured 75.3% 67.2%thinking reasoning
Humanity's Last Exam% of questions, higher is better 144 · 149 of 242 measured 11.1% 10.4%thinking reasoning
Berkeley Function Calling Leaderboard (BFCL) V4 reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 3 results
Overall accuracy · BFCL% correct, higher is better 20 · 6 of 81 measured 52.2%bfcl prompt 68.7%bfcl fc
Irrelevance detection · BFCL% correct, higher is better 41 · 4 of 81 measured 81.7%bfcl fc 95.3%bfcl prompt
Memory · BFCL% correct, higher is better 26 · 6 of 81 measured 23.9%bfcl fc 54.4%bfcl fc
UGI Leaderboard reported by UGI Leaderboard · 3 results
Writing score · UGIscore out of 100, higher is better 80 · 72 of 230 measured 49.0 52.2no reasoning
Requested-length error · UGI% off the requested word count, lower is better 69 · 17 of 230 measured 15.0% 6.0%no reasoning
Style adherence · UGIscore from 0 to 1, higher is better 64 · 22 of 230 measured 0.36 0.39thinking reasoning
Show all 25 results (13 more)
MeasureQwen3-235B-A22B-Instruct-2507Claude Haiku 4.5
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 4 results
Creative writing · TextArena rating, higher is better 122 · 112 of 385 measured 1,378 1,388
Expert prompts · TextArena rating, higher is better 103 · 97 of 338 measured 1,447 1,451
Instruction following · TextArena rating, higher is better 104 · 101 of 387 measured 1,414 1,415
Writing, literature and language · TextArena rating, higher is better 115 · 106 of 386 measured 1,396 1,400
Artificial Analysis reported by Artificial Analysis · 4 results
IFBench% of instructions, higher is better 113 · 92 of 195 measured 46.1% 54.3%thinking reasoning
Long-context reasoning (AA-LCR)% of questions, higher is better 177 · 82 of 233 measured 33.9% 74.3%thinking reasoning
Terminal-Bench Hard% of tasks, higher is better 110 · 72 of 190 measured 15.2% 27.3%no reasoning
τ²-bench telecom% of tasks, higher is better 127 · 98 of 193 measured 33.3% 54.7%thinking reasoning
Berkeley Function Calling Leaderboard (BFCL) V4 reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 5 results
Multi-turn tasks · BFCL% correct, higher is better 20 · 13 of 81 measured 45.4%bfcl fc 53.6%bfcl fc
Relevance detection · BFCL% correct, higher is better 8 · 72 of 81 measured 93.8%bfcl prompt 62.5%bfcl fc
Single-turn calls (curated) · BFCL% correct, higher is better 2 · 31 of 81 measured 90.3%bfcl prompt 86.5%bfcl fc
Single-turn calls (user-contributed) · BFCL% correct, higher is better 15 · 15 of 81 measured 78.7%bfcl prompt 78.7%bfcl fc
Web search · BFCL% correct, higher is better 20 · 2 of 81 measured 54.0%bfcl fc 83.5%bfcl fc

Quick answers

Is Qwen3-235B-A22B-Instruct-2507 better than Claude Haiku 4.5?

Pick Claude Haiku 4.5 if quality matters most: it scores higher (Intelligence Index 16.9 against 12.0). Pick Qwen3-235B-A22B-Instruct-2507 if cost matters more: it costs 6.7 times less per token.

Which is cheaper, Qwen3-235B-A22B-Instruct-2507 or Claude Haiku 4.5?

Qwen3-235B-A22B-Instruct-2507 is cheaper: $0.30 against $2.00 per million tokens, 6.7 times less.

Which is better for agents and tool use, Qwen3-235B-A22B-Instruct-2507 or Claude Haiku 4.5?

Qwen3-235B-A22B-Instruct-2507 ranks 51 of 164 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 91 for Claude Haiku 4.5.

Which is better for reasoning and knowledge, Qwen3-235B-A22B-Instruct-2507 or Claude Haiku 4.5?

Claude Haiku 4.5 ranks 127 of 198 models on sale for hard questions across science, maths and general knowledge, against 129 for Qwen3-235B-A22B-Instruct-2507.

Which is better for writing, Qwen3-235B-A22B-Instruct-2507 or Claude Haiku 4.5?

Qwen3-235B-A22B-Instruct-2507 ranks 85 of 188 models on sale for what people prefer in blind comparisons, and judged writing quality, against 87 for Claude Haiku 4.5.

Which is better for sticking to the facts, Qwen3-235B-A22B-Instruct-2507 or Claude Haiku 4.5?

Qwen3-235B-A22B-Instruct-2507 ranks 98 of 152 models on sale for summarising without inventing things, and factual answers, against 117 for Claude Haiku 4.5.

Which has the bigger context window, Qwen3-235B-A22B-Instruct-2507 or Claude Haiku 4.5?

Qwen3-235B-A22B-Instruct-2507: 262,144 tokens, against 200,000 for Claude Haiku 4.5.

Run Qwen3-235B-A22B-Instruct-2507 and Claude Haiku 4.5 on your own prompt

Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.

Try the playgroundTalk to us