Across 1 comparable benchmark, Qwen3.8-27B has better reported results on 1 and worse on 0.
Limited evidence1 of 1 shared result countedHow we compare
Price basis: Listed rates vs listed rates. The price section below shows the applicable periods.
| Fact | QwQ-Max-Preview | Qwen3.8-27B |
|---|---|---|
| Context window | 128K tokens | 256K tokens |
| Input / 1M | — | $0.500 |
| Output / 1M | — | $3.00 |
| Blended (4:1) | — | $0.712 |
| Status | Preview | Available |
| Released | March 20, 2025 | August 14, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
No shared benchmark here has a fixed 0–100 scale, so only the counted results are shown. How averages are built
Qwen3.8-27B ahead on 1 of 1 counted
No fixed-scale average — see the rows in Benchmark results
No shared comparable results for Abstract reasoning, Mathematics, Science & health, Computer & tool use, Research & search, Long context, Multimodal, Professional work & productivity, Cybersecurity.
Published scores, with matching modes where available. Gaps show QwQ-Max-Preview relative to Qwen3.8-27B. Reading the results
| Benchmark | QwQ-Max-Preview | Qwen3.8-27B | Gap |
|---|---|---|---|
| LiveCodeBenchAlgorithmic Coding · Pass @K | 65.6Normal · no tools | 90.3≠Thinking · no tools | 0.73×#63 vs #8 of 113 |
Algorithmic Coding · Pass @K
A dash means no result is available.
| Benchmark | QwQ-Max-Preview | Qwen3.8-27B |
|---|---|---|
| CritPt | — | 5.4 |
| GPQA Diamond | — | 89.2 |
| HLE | — | 30.8 |
| ProofBench v1.1 (Lean 4) | — | 16.0 |
| SciCode | — | 46.6 |
| AA Cyber Index | — | 12.7 |
| APEX-SWE | — | 50.7 |
| Code Migration | — | 14.2 |
| DeepSWE | — | 42.2 |
| IOI (Vals v2) | — | 39.1 |
| NL2Repo-Bench | — | 42.3 |
| SWE-Bench Pro - Public | — | 61.7 |
| Terminal-Bench 2.1 | — | 73.0 |
| Terminal-Bench 4.0 | — | 5.6 |
| Vibe Code Bench v1.1 | — | 64.8 |
| Agents' Last Exam v1.0 | — | 20.4 |
| APEX-Agents | — | 47.5 |
| OSWorld-Verified | — | 84.3 |
| AA Intelligence Index v5.0.1 | — | 20.1 |
| ECI | — | 149 |
| LiveBench | — | 75.3 |
| LiveBench 2.0 | — | 62.7 |
| Vals Index | — | 48.5 |
| Creative Writing | — | 1671 |
| Context Arena | — | 94.0 |
| BabyVision | — | 85.6 |
| CharXiv RQ | — | 90.2 |
| GDP.pdf | — | 16.4 |
| GeoBench ACW | — | 38.0 |
| MathVision | — | 94.6 |
| MMMU-Pro | — | 76.3 |
| OmniDocBench | — | 91.1 |
| EMB (Excel Modeling) | — | 59.7 |
| Finance Agent v2 | — | 48.5 |
| Harvey's Legal Agent Benchmark | — | 11.3 |
| Legal Research Bench | — | 36.1 |
| SAGE | — | 52.4 |
| Tax Agent Bench | — | 30.9 |
| τ³-Banking | — | 48.0 |
| ARC-AGI-1 | — | 87.5 |
| ARC-AGI-2 | — | 42.4 |
| SimpleBench | — | 60.2 |
| MedCode | — | 28.7 |
| MedScribe | — | 83.8 |
Time-of-day providers use peak-hour rates; fixed-price providers use standard rates. Off-peak discounts remain in the published details.
Drag the handle above the bar for cache hits, the one below it for the input : output mix · 5% steps
Blended price = effective input rate at 90% cache hits × 80% + output rate × 20%
A blended cost needs a published input and output price on both sides, and one of these two does not have them yet.
QwQ-Max-Preview: Listed text rates; pricing tier not confirmed.
Qwen3.8-27B: International pricing. context_length <= 1000000 Alibaba Model Studio International (Singapore)
| Price / 1M tokens | QwQ-Max-Preview | Qwen3.8-27B | Ratio |
|---|---|---|---|
| Input | — | $0.50 | — |
| Output | — | $3.00 | — |
| Cached input | — | $0.10 | — |
| Cache write | — | — | — |
| Blended | — | $0.712 | — |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
Reported cost per task, paired with the score from the same run. Settings and sources may differ; costs do not affect the verdict.
| Benchmark | QwQ-Max-Preview | Qwen3.8-27B |
|---|---|---|
| AA Intelligence Index v5.0.1 | Not reported | 1 more settings |
| ARC-AGI-1 | Not reported | 1 more settings |
| ARC-AGI-2 | Not reported | 1 more settings |
| LiveBench 2.0 | Not reported | $0.519 / task Average · Overall score (equal-weight mean of 7 suite scores, 0-100): 62.7 Max (task-specific tools) |
| Tax Agent Bench | Not reported | $0.8153 / task Average · Accuracy (%, rubric × citation quality): 30.9 Extra-High (With Tools) |
| Attribute | QwQ-Max-Preview | Qwen3.8-27B |
|---|---|---|
| Availability | Preview | Available |
| Provider | Alibaba | Alibaba |
| Model type | Reasoning model | Reasoning model |
| First release | March 20, 2025 | August 14, 2026 |
| Context window | 128K tokens | 256K tokens |
| Max output | — | 128K tokens |
| Total parameters | — | 27B |
| Active parameters | — | 27B |
| Architecture | Undisclosed | Dense |
| Documented capabilities | Reasoning, Multilingual | Reasoning, Multilingual, Coding, Tool use, Function calling, Long context, Computer use |
| Input modalities | — | Text, Image, Video |
| Output modalities | — | Text |
| Weights | Closed / API only | Open weights |
| Licence | — | Apache 2.0 |
| Commercial use | Proprietary | Commercial use allowed |
These are reported-score comparisons, not controlled head-to-head tests.
What people found comparing these models, or using either one. Comments also appear on the models' own pages.
Used QwQ-Max-Preview or Qwen3.8-27B? Sign in with Google to share what worked, what did not, and which API or setup you tested.
No community comments on either model yet.
Alternatives with the most benchmark overlap for this matchup.