Compare / Qwen3.5-122B-A10B vs Claude Haiku 4.5
Qwen3.5-122B-A10B vs Claude Haiku 4.5
Which is cheaper, which is more capable, and which does better at each kind of work, from 24 results on 4 sources that measured both models.
Of the 24 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.
Results as of 14 October 2026 (catalogue release 2026-10-14-7d7cf32c3f6d). See also Alibaba vs Anthropic, every model from both labs.
Qwen3.5-122B-A10B or Claude Haiku 4.5? The short answer
They score about the same, so pick Qwen3.5-122B-A10B: it costs 2.8 times less per token.
- Qwen3.5-122B-A10B is cheaper: $0.72 against $2.00 per million tokens, 2.8 times less.
- They score about the same on the Artificial Analysis Intelligence Index (17.7 and 16.9).
- Qwen3.5-122B-A10B is newer: released 24 Feb 2026, against 15 Oct 2025 for Claude Haiku 4.5.
- On third-party benchmarks, on the results both models have: Qwen3.5-122B-A10B does better for agents and tool use and reasoning and knowledge; Claude Haiku 4.5 for writing.
Head to head
Every fact side by side
| Qwen3.5-122B-A10B | Claude Haiku 4.5 | |
|---|---|---|
| Price per million tokens | $0.26 in · $2.08 out | $1.00 in · $5.00 out |
| Blended (3 in : 1 out) | $0.72 | $2.00 |
| Cheapest host, blended | $0.72 (Alibaba) | $2.00 (Amazon Bedrock) |
| Cached input, per million | – | $0.10 cached input |
| Intelligence Index | 17.7 | 16.9 |
| Spring Prompt overall | – | 7 |
| Output speed, tokens a second | 147 | – |
| Context window | 262,144 tokens | 200,000 tokens |
| Released | 24 Feb 2026 (newer) | 15 Oct 2025 |
| Weights | Open | Closed |
| Developer based in | China | the 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.
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.
| Measure | Qwen3.5-122B-A10B | Claude Haiku 4.5 |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results | ||
| Overall · TextArena rating, higher is better 117 · 121 of 387 measured | 1,416 | 1,414 |
| Overall · Text factualityArena rating, higher is better 103 · 118 of 163 measured | 1,428 | 1,418 |
| Business, management and finance · TextArena rating, higher is better 105 · 109 of 380 measured | 1,422 | 1,420 |
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 115 · 117 of 244 measured | 17.7no reasoning | 16.9thinking reasoning |
| GPQA Diamond% of questions, higher is better 75 · 163 of 231 measured | 85.7% | 67.2%thinking reasoning |
| Humanity's Last Exam% of questions, higher is better 93 · 150 of 243 measured | 25.2% | 10.4%thinking reasoning |
| UGI Leaderboard reported by UGI Leaderboard · 3 results | ||
| Writing score · UGIscore out of 100, higher is better 112 · 72 of 230 measured | 39.5no reasoning | 52.2no reasoning |
| Requested-length error · UGI% off the requested word count, lower is better 33 · 17 of 230 measured | 9.0%thinking reasoning | 6.0%no reasoning |
| Style adherence · UGIscore from 0 to 1, higher is better 83 · 22 of 230 measured | 0.36no reasoning | 0.39thinking reasoning |
| Vectara Hallucination Leaderboard reported by Vectara Hallucination Leaderboard · 2 results | ||
| Hallucination rate · Vectara% of summaries, lower is better 72 · 58 of 104 measured | 11.2% | 9.8% |
| Answer rate · Vectara% of documents, higher is better 31 · 50 of 104 measured | 99.8% | 99.5% |
Show all 24 results (13 more)
| Measure | Qwen3.5-122B-A10B | Claude Haiku 4.5 |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 4 results | ||
| Creative writing · TextArena rating, higher is better 136 · 112 of 385 measured | 1,366 | 1,388 |
| Expert prompts · TextArena rating, higher is better 111 · 97 of 338 measured | 1,442 | 1,451 |
| Instruction following · TextArena rating, higher is better 115 · 101 of 387 measured | 1,406 | 1,415 |
| Writing, literature and language · TextArena rating, higher is better 125 · 106 of 386 measured | 1,387 | 1,400 |
| Artificial Analysis reported by Artificial Analysis · 9 results | ||
| Terminal-Bench 2.1% of tasks, higher is better 64 · 70 of 130 measured | 47.6% | 44.2%thinking reasoning |
| Terminal-Bench 4.0% of tasks, higher is better 67 · 67 of 113 measured | 0.0% | 0.0%thinking reasoning |
| τ-bench banking% of tasks, higher is better 58 · 82 of 121 measured | 15.3% | 9.3%thinking reasoning |
| Artificial Analysis Coding Indexindex score, higher is better 65 · 68 of 131 measured | 45.7 | 43.9thinking reasoning |
| IFBench% of instructions, higher is better 22 · 92 of 195 measured | 75.7% | 54.3%thinking reasoning |
| Long-context reasoning (AA-LCR)% of questions, higher is better 73 · 83 of 234 measured | 76.3% | 74.3%thinking reasoning |
| SciCode% of problems, higher is better 80 · 74 of 114 measured | 39.7% | 42.2%thinking reasoning |
| Terminal-Bench Hard% of tasks, higher is better 65 · 72 of 190 measured | 31.1% | 27.3%no reasoning |
| τ²-bench telecom% of tasks, higher is better 30 · 98 of 193 measured | 93.6% | 54.7%thinking reasoning |
Compare them with other models
Quick answers
Is Qwen3.5-122B-A10B better than Claude Haiku 4.5?
They score about the same, so pick Qwen3.5-122B-A10B: it costs 2.8 times less per token.
Which is cheaper, Qwen3.5-122B-A10B or Claude Haiku 4.5?
Qwen3.5-122B-A10B is cheaper: $0.72 against $2.00 per million tokens, 2.8 times less.
Which is better for agents and tool use, Qwen3.5-122B-A10B or Claude Haiku 4.5?
Claude Haiku 4.5 ranks 93 of 166 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 97 for Qwen3.5-122B-A10B.
Which is better for reasoning and knowledge, Qwen3.5-122B-A10B or Claude Haiku 4.5?
Qwen3.5-122B-A10B ranks 89 of 200 models on sale for hard questions across science, maths and general knowledge, against 129 for Claude Haiku 4.5.
Which is better for writing, Qwen3.5-122B-A10B or Claude Haiku 4.5?
Claude Haiku 4.5 ranks 88 of 189 models on sale for what people prefer in blind comparisons, and judged writing quality, against 103 for Qwen3.5-122B-A10B.
Which is better for sticking to the facts, Qwen3.5-122B-A10B or Claude Haiku 4.5?
Qwen3.5-122B-A10B ranks 109 of 153 models on sale for summarising without inventing things, and factual answers, against 118 for Claude Haiku 4.5.
Which has the bigger context window, Qwen3.5-122B-A10B or Claude Haiku 4.5?
Qwen3.5-122B-A10B: 262,144 tokens, against 200,000 for Claude Haiku 4.5.
Run Qwen3.5-122B-A10B 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.