Across 12 comparable benchmarks, DeepSeek-V3.1 Terminus has better reported results on 11 and worse on 1.
Strong evidence12 of 12 shared results countedHow we compare
Price basis: Listed rates vs listed rates. The price section below shows the applicable periods.
| Fact | DeepSeek-V3.1 Terminus | Qwen3-30B-A3B-2507 |
|---|---|---|
| Context window | 128K tokens | 256K tokens |
| Input / 1M | — | — |
| Output / 1M | — | — |
| Blended (4:1) | — | — |
| Status | Available | Deprecated |
| Released | September 22, 2025 | July 29, 2025 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 6 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
DeepSeek-V3.1 Terminus ahead on 3 of 3 counted
DeepSeek-V3.1 Terminus ahead on 1 of 1 counted
DeepSeek-V3.1 Terminus ahead on 3 of 3 counted
DeepSeek-V3.1 Terminus ahead on 1 of 1 counted
Split 1–1 of 2 counted
No shared comparable results for Abstract reasoning, Computer & tool use, Research & search, Long context, Cybersecurity.
Published scores, with matching modes where available. Gaps show DeepSeek-V3.1 Terminus relative to Qwen3-30B-A3B-2507. Reading the results
| Benchmark | DeepSeek-V3.1 Terminus | Qwen3-30B-A3B-2507 | Gap |
|---|---|---|---|
| LiveCodeBenchAlgorithmic Coding · Pass @K | 74.9Normal · no tools | 43.2Normal · no tools | 1.73×#45 vs #92 of 113 |
| SWE-bench VerifiedRepository Engineering · Accuracy | 68.4Normal · no tools | 22.0≠Thinking · no tools | +46.4 pts#66 vs #107 of 108 |
| Terminal Bench HardAgentic Development · Accuracy | 31.8Normal · tools | 5.3≠Thinking · tools | +26.5 pts#65 vs #135 of 157 |
| AIME2025Mathematics · Accuracy | 88.4Thinking · no tools | 61.3≠Normal · no tools | +27.1 pts#38 vs #75 of 86 |
| CritPtScientific Reasoning · Score | 1.7Thinking · no tools | 0.3Thinking · no tools | 5.67×#92 vs #124 of 134 |
| GPQA DiamondScientific Reasoning · Accuracy | 80.7Normal · no tools | 70.4Normal · no tools | +10.3 pts#109 vs #138 of 200 |
| SciCodeScientific Computing · Score | 38.0Thinking · no tools | 33.0Thinking · no tools | 1.15×#81 vs #89 of 98 |
| GDP.pdfDocuments & Charts · Task pass rate (%) | 5.4Thinking · no tools | 0.4Thinking · no tools | +5.0 pts#73 vs #96 of 96 |
| τ²-BenchService Workflows · Accuracy | 37.0Normal · tools | 49.0≠Thinking · tools | −12.0 pts#37 vs #33 of 40 |
| τ³-BankingService Workflows · Score | 21.0Thinking · tools | 5.4Thinking · tools | 3.89×#56 vs #107 of 120 |
| HLEKnowledge Exams · Accuracy | 21.7Normal · no tools | 9.8≠Thinking · no tools | +11.9 pts#97 vs #128 of 151 |
| MMLU-ProKnowledge Exams · Accuracy | 85.0Normal · no tools | 78.4Normal · no tools | +6.6 pts#31 vs #69 of 153 |
Algorithmic Coding · Pass @K
Repository Engineering · Accuracy
Agentic Development · Accuracy
Mathematics · Accuracy
Scientific Reasoning · Score
Scientific Reasoning · Accuracy
Scientific Computing · Score
Documents & Charts · Task pass rate (%)
Service Workflows · Accuracy
Service Workflows · Score
A dash means no result is available.
| Benchmark | DeepSeek-V3.1 Terminus | Qwen3-30B-A3B-2507 |
|---|---|---|
| Terminal-Bench 1.0 | 36.7 | — |
| BrowseComp | — | 2.3 |
| SimpleQA | 96.8 | — |
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.
DeepSeek-V3.1 Terminus: Listed text rates; pricing tier not confirmed.
Qwen3-30B-A3B-2507: Listed text rates; pricing tier not confirmed.
| Price / 1M tokens | DeepSeek-V3.1 Terminus | Qwen3-30B-A3B-2507 | Ratio |
|---|---|---|---|
| Input | — | — | — |
| Output | — | — | — |
| Cached input | — | — | — |
| Cache write | — | — | — |
| Blended | — | — | — |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
Lower prices are highlighted. The cache hit rate applies to input tokens only; models without a published cached-input rate use their normal input price for those tokens.| Attribute | DeepSeek-V3.1 Terminus | Qwen3-30B-A3B-2507 |
|---|---|---|
| Availability | Available | Deprecated |
| Provider | DeepSeek-AI | Alibaba |
| Model type | Chat model | Chat model |
| First release | September 22, 2025 | July 29, 2025 |
| Context window | 128K tokens | 256K tokens |
| Max output | 8K tokens | 16K tokens |
| Total parameters | 671B | 30.5B |
| Active parameters | 37B | 3.3B |
| Architecture | Mixture of Experts (MoE) | Mixture of Experts (MoE) |
| Documented capabilities | Reasoning, Multilingual | Multilingual |
| Input modalities | Text | Text |
| Output modalities | Text | Text |
| Weights | Open weights | Open weights |
| Licence | MIT License | Apache 2.0 |
| 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 DeepSeek-V3.1 Terminus or Qwen3-30B-A3B-2507? 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.