AlibabaQwen3.5 Flash
Qwen3.5 Flash is an Alibaba proprietary reasoning model, released 25 Feb 2026. Its best configuration ranks #330 of 785 on the BenchLeader Index at 52.0 ±5.8, in the upper half. The ± is the point: 346 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in reasoning (59) and lowest in coding (47). At $0.175 per million tokens blended it is among the cheapest fifth of ranked models. Output speed of 86 tokens per second puts it slower than most, with a first token in 0.6 s. It has been measured at 2 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 22 Oct 2026.
- Blended price
- $0.175/M
- $0.100 in · $0.400 out
- Output speed
- 86 tok/s
- OpenRouter traffic, 7-day median; not yet measured by Artificial Analysis
- First answer
- 0.57 s
- Context
- 1M
- Full answer
- –
- Cost per run
- –
- Released
- 25 Feb 2026
- Overall index52
- Reasoning59
- Coding47
- Maths50
- Knowledge50
- Human preference57
Versions
Alibaba has shipped 4 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| Qwen3.8 Flash | 26 Aug 2026 | – | – |
| Qwen3.7 Flashno reasoning | 27 Jul 2026 | – | – |
| Qwen3.6 Flash | 27 Apr 2026 | 44.1 | #550 |
| Qwen3.5 Flashthis page | 25 Feb 2026 | 52.0 | #330 |
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| no reasoning | – | – | 86 tok/s | 0.57 s | $0.0002 | Maths 35 |
| not statedbest | 52.0 | #330 | 86 tok/s | 0.57 s | $0.0002 | Coding 47 · Human preference 57 · Knowledge 50 · Maths 50 · Reasoning 59 |
How its index has moved
8 Sept 2026 to 21 Oct 2026The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | no reasoning | not stated | Source | Trend |
|---|---|---|---|---|
| GPQA Diamondeffort not stated | – | 82.3%#113 | Epoch AI Benchmarking Hub | |
| Arena Hard Promptseffort not stated | – | 1416#157 | Arena (LMArena) | |
| GPQA Diamond (Vals)not in indexeffort not stated | – | 82.8%#58 | Vals AI | |
| Chess Puzzlesnot in indexeffort not stated | – | 14.0%#111 | Epoch AI Benchmarking Hub | |
| Mystery Game Puzzlesnot in index | 16.0%#84 | 20.0%#60 | Epoch AI Benchmarking Hub | |
| LMCAnot in indexeffort not stated | – | 29.1%#140 | Epoch AI Benchmarking Hub | |
| DTBenchnot in indexeffort not stated | – | 82.9%#116 | Epoch AI Benchmarking Hub |
Coding
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Coding | 1436#162 | Arena (LMArena) | |
| Arena WebDev | 1244#127 | Arena (LMArena) | |
| LiveCodeBench | 83.3%#47 | Vals AI | |
| SWE-bench (Vals)not in index | 64.4%#71 | Vals AI | |
| ALE-Benchnot in index | 221.8#122 | Epoch AI Benchmarking Hub |
Agents & tools
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Terminal-Bench 2.0 (Vals)not in index | 24.7%#51 | Vals AI | |
| Vending-Bench 2not in index | 462.7#59 | Andon Labs |
Maths
| Benchmark | no reasoning | not stated | Source | Trend |
|---|---|---|---|---|
| FrontierMath Tiers 1–3 | 18.3%#95 | 9.5%#106 | Epoch AI Benchmarking Hub | |
| OTIS Mock AIMEeffort not stated | – | 84.4%#98 | Epoch AI Benchmarking Hub | |
| AIME (Vals)effort not stated | – | 92.5%#19 | Vals AI | |
| Arena Mathseffort not stated | – | 1402#148 | Arena (LMArena) |
Knowledge
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| SimpleQA Verified | 20.3%#73 | Epoch AI Benchmarking Hub | |
| MMLU-Pro | 84.1%#65 | Vals AI | |
| LegalBench | 84.3%#36 | Vals AI | |
| CorpFin | 63.6%#44 | Vals AI | |
| TaxEval | 72.2%#66 | Vals AI | |
| CaseLaw v2not in index | 56.0%#31 | Vals AI | |
| MedCodenot in index | 33.0%#83 | Vals AI | |
| MedScribenot in index | 70.6%#90 | Vals AI |
Instruction following
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Instruction Followingnot in index | 1382#160 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1396#153 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1338#178 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1396#159 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1400#160 | Arena (LMArena) |
Multimodal
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| MMMU-Pro (Vals)not in index | 81.9%#38 | Vals AI | |
| MortgageTaxnot in index | 67.4%#25 | Vals AI | |
| SAGEnot in index | 42.5%#53 | Vals AI |
Composite
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Epoch Capabilities Indexnot in index | 143.5#95 | Epoch AI Benchmarking Hub |
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| Alibaba Cloud Int. | 67 tok/s | 0.56 s | $0.065 | $0.260 | 1M | – |
Price history
The price above, per 1M tokens, every day it changed.
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.010 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0002 | $0.0001 | 4.1 s |
| Summarize a 30-page report | 12,000 / 600 | $0.0014 | $0.0006 | 7.5 s |
| Code edit | 6,000 / 1,500 | $0.0012 | $0.0008 | 18.0 s |
| Agentic coding session | 60,000 / 4,000 | $0.0076 | $0.0036 | 47.1 s |
| Structured extraction | 2,000 / 200 | $0.0003 | $0.0001 | 2.9 s |
More from Alibaba
See also
Questions people ask
- What are Qwen3.5 Flash's benchmark scores?
- As of 24 Oct 2026, Qwen3.5 Flash scores 52.0 ±5.8 on the BenchLeader Index, strongest in reasoning (59) and weakest in coding (47). That rests on 15 independent benchmark results, and every score on this page links to its publisher.
- How does Qwen3.5 Flash rank?
- Qwen3.5 Flash is #330 of 785 ranked configurations on 24 Oct 2026. 346 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- How much does Qwen3.5 Flash cost per 1M tokens?
- As of 24 Oct 2026, Qwen3.5 Flash costs $0.100 per 1M input tokens and $0.400 per 1M output tokens, $0.175 blended (three input tokens per output token). The list price last changed on 24 Oct 2026.
- When was Qwen3.5 Flash released?
- Alibaba released Qwen3.5 Flash on 25 Feb 2026. BenchLeader has results for it at 2 reasoning-effort settings.
Data as of 24 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “Qwen3.5 Flash: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/qwen3-5-flash.