AlibabaQwen3.6 27B
Qwen3.6 27B is an Alibaba open-weights reasoning model, released 22 Apr 2026. Its best configuration (thinking reasoning effort) ranks #266 of 785 on the BenchLeader Index at 53.9 ±5.8, in the upper half. The ± is the point: 331 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 instruction following (67) and lowest in reasoning (43). At $1.35 per million tokens blended it is pricier than most ranked models. Output speed of 58 tokens per second puts it slower than most, with a first answer in 102.2 s. It has been measured at 3 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 22 Apr 2026.stale: no new result in six months
- Blended price
- $1.35/M
- $0.600 in · $3.60 out
- Output speed
- 58 tok/s
- measured by Artificial Analysis
- First answer
- 102 s
- first token 0.53 s
- Context
- 262k
- Full answer
- 111 s
- median, reasoning included
- Cost per run
- $0.621
- one full Intelligence Index run
- Released
- 22 Apr 2026
- Overall index54
- Reasoning43
- Agents & tools51
- Knowledge53
- Instruction following67
- Multimodal58
- Long context64
- Composite53
Versions
Alibaba has shipped 3 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 27Bxhigh | 14 Aug 2026 | 56.2 | #211 |
| Qwen3.6 27Bthinkingthis page | 22 Apr 2026 | 53.9 | #266 |
| Qwen3.5-27Bthinking | 24 Feb 2026 | 54.7 | #246 |
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 | 47.9 | #446 | 58 tok/s | 3.62 s | $0.0013 | Agents & tools 42 · Coding 36 · Composite 52 · Instruction following 48 · Knowledge 38 · Long context 58 · Maths 47 · Multimodal 55 · Reasoning 52 |
| thinkingbest | 53.9 | #266 | 58 tok/s | 102 s | $0.0013 | Agents & tools 51 · Composite 53 · Instruction following 67 · Knowledge 53 · Long context 64 · Multimodal 58 · Reasoning 43 |
| not stated | 48.6 | #422 | 57 tok/s | 0.53 s | $0.0013 | Human preference 24 · Knowledge 55 · Maths 51 · Reasoning 62 |
How its index has moved
9 Sept 2026 to 21 Oct 2026- thinking
- not stated
- no reasoning
The 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 | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| GPQA Diamond | 84.8%#93 | – | 85.9%#84 | Epoch AI Benchmarking Hub | |
| LiveBench 1.0 Reasoningnot in indexeffort not stated | – | – | 70.3%#65 | LiveBench | |
| GPQA Diamond (AA)not in index | 82.9%#163 | 84.2%#145 | – | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 15.1%#252 | 23.1%#196 | – | Artificial Analysis | |
| CritPt | 0.9%#237 | 1.1%#223 | – | Artificial Analysis | |
| Mystery Game Puzzlesnot in index | 7.0%#123 | – | – | Epoch AI Benchmarking Hub | |
| LMCAnot in indexeffort not stated | – | – | 34.5%#127 | Epoch AI Benchmarking Hub | |
| DTBenchnot in indexeffort not stated | – | – | 78.1%#140 | Epoch AI Benchmarking Hub |
Coding
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| SciCode | 37.3%#180 | – | – | SciCode | |
| LiveBench 1.0 Codingnot in indexeffort not stated | – | – | 71.8%#58 | LiveBench | |
| SciCode (AA)not in index | – | 42.8%#158 | – | Artificial Analysis | |
| SWE-bench (Vals)not in indexeffort not stated | – | – | 70.0%#58 | Vals AI | |
| Vibe Code Bench v1.1not in indexeffort not stated | – | – | 11.9%#94 | Vals AI | |
| Terminal-Bench 4.0 (AA)not in index | – | 0.0%#152 | – | Artificial Analysis | |
| Terminal-Bench 2.1 (AA)not in index | 51.3%#117 | 60.7%#98 | – | Artificial Analysis |
Agents & tools
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| LiveBench Agentic Codingnot in indexeffort not stated | – | – | 39.3%#64 | LiveBench | |
| Terminal-Bench Hard | 21.2%#158 | 34.9%#82 | – | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 93.6%#41 | 94.2%#33 | – | Artificial Analysis | |
| τ³-Banking (AA) | 9.3%#150 | 16.7%#109 | – | Artificial Analysis | |
| Terminal-Bench 2.0 (Vals)not in indexeffort not stated | – | – | 44.9%#24 | Vals AI | |
| AutomationBenchnot in index | – | 11.1%#154 | – | Artificial Analysis | |
| GDP.pdf (AA)not in index | – | 11.0%#126 | – | Artificial Analysis |
Maths
| Benchmark | no reasoning | not stated | Source | Trend |
|---|---|---|---|---|
| FrontierMath Tiers 1–3 | 34.0%#70 | 35.1%#67 | Epoch AI Benchmarking Hub | |
| OTIS Mock AIME | 66.7%#149 | 91.1%#70 | Epoch AI Benchmarking Hub | |
| LiveBench Mathematicsnot in indexeffort not stated | – | 79.9%#60 | LiveBench |
Knowledge
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| LiveBench 1.0 Data Analysisnot in indexeffort not stated | – | – | 70.4%#50 | LiveBench | |
| AA-Omniscience | -53.2#417 | -20.0#244 | – | Artificial Analysis | |
| CorpFineffort not stated | – | – | 62.3%#48 | Vals AI | |
| TaxEvaleffort not stated | – | – | 71.3%#76 | Vals AI | |
| CaseLaw v2not in indexeffort not stated | – | – | 53.2%#40 | Vals AI | |
| AA-Omniscience: accuracynot in index | 16.7%#382 | 19.6%#312 | – | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 16.1%#329 | 50.7%#113 | – | Artificial Analysis |
Instruction following
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| LiveBench Languagenot in indexeffort not stated | – | – | 63.3%#63 | LiveBench | |
| LiveBench Instruction Followingnot in indexeffort not stated | – | – | 53.2%#66 | LiveBench | |
| IFBench | 45.7%#201 | 67.5%#85 | – | Artificial Analysis |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| EQ-Bench 4 | 1026#26 | EQ-Bench |
Multimodal
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| MMMU-Pro | 71.7%#136 | 74.6%#109 | – | Artificial Analysis | |
| MortgageTaxnot in indexeffort not stated | – | – | 68.3%#18 | Vals AI | |
| SAGEnot in indexeffort not stated | – | – | 45.6%#39 | Vals AI |
Long context
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| AA-LCR | 66.7%#245 | 77.3%#132 | Artificial Analysis | |
| MLCRnot in index | – | 10.6%#108 | Artificial Analysis |
Composite
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| Epoch Capabilities Indexnot in indexeffort not stated | – | – | 146.2#76 | Epoch AI Benchmarking Hub | |
| LiveBench 1.0not in indexeffort not stated | – | – | 64.0%#64 | LiveBench | |
| AA Intelligence Index v4.3.2 | 19.8#237 | 21.4#215 | – | Artificial Analysis |
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 |
|---|---|---|---|---|---|---|
| DeepInfra | 65 tok/s | 0.68 s | $0.320 | $3.20 | 262k | fp8 |
| Reka AI | 60 tok/s | 0.48 s | $0.200 | $2.00 | 262k | – |
| Venice | 52 tok/s | 0.53 s | $0.325 | $3.25 | 256k | fp8 |
| Alibaba Cloud Int. | 45 tok/s | 0.83 s | $0.450 | $2.70 | 262k | – |
| Chutes | 29 tok/s | 2.19 s | $0.250 | $2.00 | 262k | fp8 |
| SiliconFlow | 24 tok/s | 1.06 s | $0.300 | $3.20 | 262k | fp8 |
| Phala | 23 tok/s | 1.58 s | $0.320 | $3.25 | 262k | – |
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 $0.372 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0013 | $0.0013 | 1.8 min |
| Summarize a 30-page report | 12,000 / 600 | $0.0094 | $0.0073 | 1.9 min |
| Code edit | 6,000 / 1,500 | $0.0090 | $0.0080 | 2.1 min |
| Agentic coding session | 60,000 / 4,000 | $0.050 | $0.040 | 2.9 min |
| Structured extraction | 2,000 / 200 | $0.0019 | $0.0016 | 1.8 min |
See also
Questions people ask
- What are Qwen3.6 27B's benchmark scores?
- As of 24 Oct 2026, Qwen3.6 27B (thinking effort) scores 53.9 ±5.8 on the BenchLeader Index, strongest in instruction following (67) and weakest in reasoning (43). That rests on 8 independent benchmark results, and every score on this page links to its publisher.
- How does Qwen3.6 27B rank?
- Qwen3.6 27B (thinking effort) is #266 of 785 ranked configurations on 24 Oct 2026. 331 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 Qwen3.6 27B cost per 1M tokens?
- As of 24 Oct 2026, Qwen3.6 27B costs $0.600 per 1M input tokens and $3.60 per 1M output tokens, $1.35 blended (three input tokens per output token).
- When was Qwen3.6 27B released?
- Alibaba released Qwen3.6 27B on 22 Apr 2026. BenchLeader has results for it at 3 reasoning-effort settings.
Data as of 24 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “Qwen3.6 27B: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/qwen3-6-27b.