Models / Qwen3.5-35B-A3B

Alibaba

Qwen3.5-35B-A3B

Qwen3.5-35B-A3B trails most of the field on what it has been measured on; worth a look for simple, high-volume work where cost matters most. The cheapest model of similar intelligence is Gemma 4 26B A4B, at $0.14 per million tokens against $0.36.

Superseded. Qwen3.6-35B-A3B is Alibaba's newer model in this line, at $0.31 blended: see Qwen3.6-35B-A3B or compare the two.

Results 16 Nov 2026 · 32 results · 4 sources

Test Qwen3.5-35B-A3B on your prompt → It runs beside the models you could use instead, with every answer checked and costed.

Price per million tokens
$0.15 in · $1.00 out
Typical host on OpenRouter, 10 Nov 2026 · compare prices
Speed
Not measured
Intelligence Index
19.3
Artificial Analysis; 15.1–19.3 across 2 settings
Spring Prompt overall
Not ranked yet
needs our benchmarks and two groups of results
Developer
Alibaba, based in China
Weights published
Released
24 Feb 2026 (Artificial Analysis)
262,144 tokens of context

Where it stands among the models you could choose

Available models grouped by price, intelligence, speed and developer location, plus models with published weights. Retired models are excluded.

AmongIntelligence IndexSpring Prompt overallPrice, blendedSpeed
Models available today 101st of 185Claude Opus 5.5 – $0.36 blended, 30th percentilecheapest: Mistral Nemo –
Models priced $0.25–1 per million tokens 25th of 61MiMo-V2.6-Pro – sets this group –
Models with similar intelligence sets this group – $0.36 blended, 21st percentilecheapest: Gemma 4 26B A4B –
Models from developers based in China 41st of 75MiMo-V2.6-Pro – $0.36 blended, 32nd percentilecheapest: Qwen3.5-9B –
Open-weights models 41st of 104MiMo-V2.6-Pro – $0.36 blended, 45th percentilecheapest: Mistral Nemo –

The name under each rank is the leader of that group; green is the top quarter of the group and red the bottom quarter, by rank. Price is shown as a percentile: the share of the group that costs less, so lower is cheaper. "Sets this group" marks the measure the group is defined by. Spring Prompt overall is our score out of 100 across every source (how it works). Price is the developer's list price (or the typical OpenRouter host where we haven't read one) for three input tokens to one output; intelligence and speed are from Artificial Analysis (data sourced from Artificial Analysis); "similar intelligence" means within 4 points on its Intelligence Index. Developer locations are where each company is based, not where a model is served. Groups under 3 models are not ranked.

How good is it, and for what?

Each use case ranks the available models its benchmarks measured, then those priced $0.25–1 per million tokens. The bar shows its percentile in that field, best to the right. Open a row for the results behind it.

Product listingsTurning a sparse product feed and photos into listings that can go live Not measured
  • CatalogBench: has not measured this model.
Decks from an analysisTurning a finished analysis into a deck you could present as it is Not measured
  • DeckBench: has not measured this model.
User surveysPlanning a user survey and reading its results without being misled Not measured
  • SurveyBench: has not measured this model.
Marketing planningPlanning a year of ad spend without overspending Not measured
  • ROASBench: has not measured this model.
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 121st of 159 available · 32nd of 47 at its price
BenchmarkQwen3.5-35B-A3BRank among availableBest available
Artificial AnalysisArtificial Analysis: Terminal-Bench 2.1 · reported 40.8% 69th of 107 Claude Fable 5.1 91.4%
Artificial AnalysisArtificial Analysis: τ-bench banking · reported 4.9% 92nd of 104 Grok 4.6 50.7%
  • tau2-bench: has not measured this model.
  • Berkeley Function Calling Leaderboard (BFCL) V4: pending. Its latest results are dated 16 Dec 2025, before this model was released.
  • OpenHands Index: has not measured this model.
  • Microsoft STATE-Bench: has not measured this model.
  • Vending-Bench 2: has not measured this model.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects Not measured
  • APEX-Agents: has not measured this model.
  • GDP.pdf: has not measured this model.
  • Remote Labor Index: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge 97th of 187 available · 23rd of 61 at its price
BenchmarkQwen3.5-35B-A3BRank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 19.3 101st of 185 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 21.0% 100th of 186 Claude Opus 5.5 61.4%
Artificial AnalysisArtificial Analysis: GPQA Diamond · reported 84.5% 73rd of 168 GPT-6 Astra 96.3%
WritingWhat people prefer in blind comparisons, and judged writing quality 100th of 169 available · 28th of 58 at its price
BenchmarkQwen3.5-35B-A3BRank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,394 101st of 153 Gemini 4 Argon 1,525
UGI LeaderboardUGI: writing score · reported 42.6 71st of 124 Gemini 3.8 Flash 78.6
Sticking to the factsSummarising without inventing things, and factual answers 108th of 139 available · 23rd of 40 at its price
BenchmarkQwen3.5-35B-A3BRank among availableBest available
Vectara Hallucination LeaderboardVectara: hallucination rate · reported 10.5% 43rd of 71 GPT-5.4 nano 3.1%
Arena (formerly LMArena)Arena Text factuality: overall · reported 1,407 94th of 109 Gemini 4 Argon 1,511
  • SimpleQA Verified (Epoch AI): has not measured this model.
Speed under pressureGood decisions against a real clock (fast chess) Not measured
  • BulletBench: has not measured this model.

Nearest alternatives

Against Qwen3.5-35B-A3B at $0.36 blended, Intelligence Index 19.3. "Similar intelligence" means within 4 points or better.

Where to run it

7 hosts on OpenRouter at the standard service tier, with a median of $0.36 blended per million tokens; quantisation where the host states it: FP4, FP8. Cheapest first.

HostQuantisationInputOutputBlended
DarkbloomFP4$0.07$0.75$0.24
DeepInfraFP8$0.14$1.00$0.35
ParasailFP8$0.15$1.00$0.36
Venicenot stated$0.15$1.00$0.36
Alibabanot stated$0.16$1.30$0.45
AtlasCloudFP8$0.23$1.80$0.62
SiliconFlowFP8$0.24$1.80$0.63

Prices per million tokens from OpenRouter's public endpoint list, checked 10 Nov 2026.

Against the alternatives

The models you would most likely weigh it against: the leaders of the groups above.

Scroll sideways to see every alternative.

Qwen3.5-35B-A3BMiMo-V2.6-Pro
most intelligent at $0.25–1 per million tokens
Gemma 4 26B A4B
cheapest with similar intelligence
Claude Opus 5.5
most intelligent available today
Price per million tokens $0.36 $0.54$0.14$8.00
Tokens a second – 438296
Intelligence Index 19.3 46.316.757.6
Spring Prompt overall – ––82
Rank among available models, by use case
Product listings – ––13th
Decks from an analysis – ––7th
User surveys – ––4th
Marketing planning – ––9th
Agents and tool use 121st of 159 23rd105th3rd
Professional work – ––3rd
Reasoning and knowledge 97th of 187 13th110th1st
Writing 100th of 169 23rd82nd5th
Sticking to the facts 108th of 139 10th43rd3rd
Speed under pressure – –––

Shaded figures are better than Qwen3.5-35B-A3B's. A dash means no result.

Where it has been measured

25th most intelligent of the 61 models priced $0.25–1 per million tokens. Among models available today, in the bottom quarter for agents and tool use.

Artificial Analysis lists it but has not published a speed for it yet

Every result

32 published results from 4 sources, each in the source's own units, with the configuration that produced it.

Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 7 results
MeasureValueRankConfigurationDated
Arena Text factuality: overallArena rating, higher is better 1,407
1,404–1,411
148 of 182 Qwen3.5-35B-A3B 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,403
1,395–1,411
159 of 406 Qwen3.5-35B-A3B 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,342
1,333–1,351
186 of 411 Qwen3.5-35B-A3B 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,419
1,408–1,431
154 of 363 Qwen3.5-35B-A3B 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,386
1,379–1,392
162 of 413 Qwen3.5-35B-A3B 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,394
1,390–1,398
168 of 413 Qwen3.5-35B-A3B 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,363
1,356–1,371
178 of 412 Qwen3.5-35B-A3B 2 Oct 2026

Not shown: Not an error rate: claims that cannot be checked on the web are skipped, and preference still carries most of the weight.

Artificial Analysis · reported by Artificial Analysis · 17 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Coding Indexindex score, higher is better 37.0 127 of 187 Qwen3.5-35B-A3B (no reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 19.3 217 of 413 Qwen3.5-35B-A3B 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 15.1 255 of 413 Qwen3.5-35B-A3B (no reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 84.5% 127 of 360 Qwen3.5-35B-A3B 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 81.9% 160 of 360 Qwen3.5-35B-A3B (no reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 21.0% 196 of 419 Qwen3.5-35B-A3B 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 13.4% 247 of 419 Qwen3.5-35B-A3B (no reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 72.5% 43 of 288 Qwen3.5-35B-A3B 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 44.5% 171 of 288 Qwen3.5-35B-A3B (no reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 72.0% 182 of 407 Qwen3.5-35B-A3B 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 63.0% 244 of 407 Qwen3.5-35B-A3B (no reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 2.1% of tasks, higher is better 40.8% 122 of 188 Qwen3.5-35B-A3B (no reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 26.5% 121 of 282 Qwen3.5-35B-A3B 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 10.6% 198 of 282 Qwen3.5-35B-A3B (no reasoning) 20 Oct 2026
Artificial Analysis: τ-bench banking% of tasks, higher is better 4.9% 159 of 175 Qwen3.5-35B-A3B (no reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 89.2% 59 of 286 Qwen3.5-35B-A3B 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 86.3% 72 of 286 Qwen3.5-35B-A3B (no reasoning) 20 Oct 2026

Not shown: Not your codebase or tools.

UGI Leaderboard · reported by UGI Leaderboard · 6 results
MeasureValueRankConfigurationDated
UGI: requested-length error% off the requested word count, lower is better 18.0% 158 of 379 Qwen3.5-35B-A3B (think-prefill reasoning) 14 Mar 2026
UGI: requested-length error% off the requested word count, lower is better 17.0% 148 of 379 Qwen3.5-35B-A3B (no reasoning) 14 Mar 2026
UGI: style adherencescore from 0 to 1, higher is better 0.28 366 of 379 Qwen3.5-35B-A3B (think-prefill reasoning) 14 Mar 2026
UGI: style adherencescore from 0 to 1, higher is better 0.33 263 of 379 Qwen3.5-35B-A3B (no reasoning) 14 Mar 2026
UGI: writing scorescore out of 100, higher is better 42.6 202 of 379 Qwen3.5-35B-A3B (think-prefill reasoning) 14 Mar 2026
UGI: writing scorescore out of 100, higher is better 37.0 240 of 379 Qwen3.5-35B-A3B (no reasoning) 14 Mar 2026

Not shown: Not other format limits such as character counts or bullet counts.

Vectara Hallucination Leaderboard · reported by Vectara Hallucination Leaderboard · 2 results
MeasureValueRankConfigurationDated
Vectara: answer rate% of documents, higher is better 99.8% 34 of 108 Qwen3.5-35B-A3B 22 Sep 2026
Vectara: hallucination rate% of summaries, lower is better 10.5% 62 of 108 Qwen3.5-35B-A3B 22 Sep 2026

Not shown: Not a quality score: a low rate usually means content filters were triggered, and hallucination rates are measured on answered documents only.

Will Qwen3.5-35B-A3B work for your product?

Benchmarks aren't your task. Paste the prompt your product uses and run it beside the models you could use instead, with every answer checked and costed.

New here? Your email starts a free trial. No credit card needed.