Across 7 comparable benchmarks, DeepSeek-R1-0528 has better reported results on 7 and worse on 0.
Moderate evidence7 of 7 shared results countedHow we compare
Price basis: Listed rates vs standard rates. The price section below shows the applicable periods.
| Fact | QwQ-32B | DeepSeek-R1-0528 |
|---|---|---|
| Context window | 128K tokens | 64K tokens |
| Input / 1M | — | $0.500 |
| Output / 1M | — | $2.15 |
| Blended (4:1) | — | $0.722 |
| Status | Available | Available |
| Released | March 6, 2025 | May 28, 2025 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 4 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
DeepSeek-R1-0528 ahead on 1 of 1 counted
DeepSeek-R1-0528 ahead on 2 of 2 counted
DeepSeek-R1-0528 ahead on 1 of 1 counted
DeepSeek-R1-0528 ahead on 1 of 1 counted
No fixed-scale average — see the rows in Benchmark results
No shared comparable results for Abstract reasoning, Computer & tool use, Research & search, Long context, Multimodal, Cybersecurity.
Published scores, with matching modes where available. Gaps show QwQ-32B relative to DeepSeek-R1-0528. Reading the results
| Benchmark | QwQ-32B | DeepSeek-R1-0528 | Gap |
|---|---|---|---|
| Aider-PolyglotAgentic Development · Accuracy | 20.9Thinking · no tools | 71.4Thinking · no tools | −50.5 pts#38 vs #11 of 47 |
| AIME 2024Mathematics · Accuracy | 79.5Normal · no tools | 91.4≠Thinking · no tools | −11.9 pts#25 vs #10 of 54 |
| MATH-500Mathematics · Accuracy | 91.0Normal · no tools | 98.0≠Thinking · no tools | −7.0 pts#41 vs #16 of 49 |
| GPQA DiamondScientific Reasoning · Accuracy | 58.0Normal · no tools | 81.0≠Thinking · no tools | −23.0 pts#168 vs #106 of 200 |
| Creative WritingWriting | 1258Normal · no tools | 1422Normal · no tools | 0.89×#90 vs #70 of 121 |
| MMLU-ProKnowledge Exams · Accuracy | 76.0Normal · no tools | 85.0≠Thinking · no tools | −9.0 pts#79 vs #31 of 153 |
| ECICapability Indices · ECI score (capability index, higher is better) | 138Reported best | 141Reported best | 0.97×#107 vs #93 of 168 |
Agentic Development · Accuracy
Scientific Reasoning · Accuracy
Writing
A dash means no result is available.
| Benchmark | QwQ-32B | DeepSeek-R1-0528 |
|---|---|---|
| AIME2025 | — | 87.5 |
| CritPt | — | 1.4 |
| HLE | — | 17.7 |
| IMO-ProofBench | — | 29.0 |
| IMO-ProofBench Advanced | — | 3.8 |
| HumanEval | 19.0 | — |
| LiveCodeBench | — | 73.3 |
| SWE-bench Verified | — | 57.6 |
| Terminal Bench Hard | — | 15.9 |
| Terminal-Bench 1.0 | — | 5.7 |
| ARC-AGI-1 | — | 21.2 |
| ARC-AGI-2 | — | 1.3 |
| SimpleBench | — | 40.8 |
| SimpleQA | — | 27.8 |
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-32B: Listed text rates; pricing tier not confirmed.
DeepSeek-R1-0528: Standard pricing. DeepInfra official models API (deepseek-ai/DeepSeek-R1-0528) verified 2026-11-13
| Price / 1M tokens | QwQ-32B | DeepSeek-R1-0528 | Ratio |
|---|---|---|---|
| Input | — | $0.50 | — |
| Output | — | $2.15 | — |
| Cached input | — | $0.35 | — |
| Cache write | — | — | — |
| Blended | — | $0.722 | — |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
| Attribute | QwQ-32B | DeepSeek-R1-0528 |
|---|---|---|
| Availability | Available | Available |
| Provider | Alibaba | DeepSeek-AI |
| Model type | Reasoning model | Reasoning model |
| First release | March 6, 2025 | May 28, 2025 |
| Context window | 128K tokens | 64K tokens |
| Max output | — | 64K tokens |
| Total parameters | 32.5B | 671B |
| Active parameters | — | 37B |
| Architecture | Dense | Mixture of Experts (MoE) |
| Documented capabilities | Reasoning, Multilingual | Reasoning, Multilingual |
| Input modalities | Text | Text |
| Output modalities | Text | Text |
| Weights | Open weights | Open weights |
| Licence | Apache 2.0 | MIT License |
| Commercial use | Commercial use allowed | 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-32B or DeepSeek-R1-0528? 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.