Model comparison
DeepSeek V4 Pro (Max) vs Qwen3.5 397B
Head-to-head evidence from 26 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: DeepSeek V4 Pro (Max) unranked (Not scored); Qwen3.5 397B #71 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. DeepSeek V4 Pro (Max) and Qwen3.5 397B share 26 comparable benchmark results. 4 of 8 categories are comparable. 22 results are unique to DeepSeek V4 Pro (Max); 29 to Qwen3.5 397B.
Updated July 23, 2026- Shared results
- 26
- DeepSeek V4 Pro (Max) only
- 22
- Qwen3.5 397B only
- 29
- Comparable categories
- 4 / 8
Treat this as a split decision. DeepSeek V4 Pro (Max) makes more sense if agentic is the priority or you want the cheaper token bill; Qwen3.5 397B is the better fit if you would rather avoid the extra latency and token burn of a reasoning model.
Confidence note. This is a partial-evidence comparison with 26 shared benchmark results across 6 evidence categories; 4 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
DeepSeek V4 Pro (Max) and Qwen3.5 397B finish on the same BenchAlign overall score, so this is less about a single winner and more about where the edge shows up. The BenchAlign headline says tie; the benchmark table is where the real choice happens.
Qwen3.5 397B is also the more expensive model on tokens at $0.60 input / $3.60 output per 1M tokens, versus $0.43 input / $0.87 output per 1M tokens for DeepSeek V4 Pro (Max). That is roughly 4.1x on output cost alone. DeepSeek V4 Pro (Max) is the reasoning model in the pair, while Qwen3.5 397B is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. DeepSeek V4 Pro (Max) gives you the larger context window at 1M, compared with 128K for Qwen3.5 397B.
Category breakdown
Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.
| Category | DeepSeek V4 Pro (Max) | Δ | Qwen3.5 397B |
|---|---|---|---|
| Agentic | DeepSeek V4 Pro (Max)74.5 | Margin← 18.0 | Qwen3.5 397B56.5 |
| Math | DeepSeek V4 Pro (Max)95.2 | Margin← 4.6 | Qwen3.5 397B90.6 |
| Coding | DeepSeek V4 Pro (Max)70.9 | Margin← 4.4 | Qwen3.5 397B66.5 |
| Knowledge | DeepSeek V4 Pro (Max)60.1 | Margin← 3.5 | Qwen3.5 397B56.6 |
| Reasoning | DeepSeek V4 Pro (Max)Not measured | MarginNo overlap | Qwen3.5 397B63.2 |
| Multilingual | DeepSeek V4 Pro (Max)Not measured | MarginNo overlap | Qwen3.5 397B84.7 |
| Multimodal | DeepSeek V4 Pro (Max)Not measured | MarginNo overlap | Qwen3.5 397B79.6 |
| Inst. Following | DeepSeek V4 Pro (Max)Not measured | MarginNo overlap | Qwen3.5 397B92.6 |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
BrowseComp
AgenticA 83.4%B 62%Winner: DeepSeek V4 Pro (Max)Δ 21.4BrowseComp: DeepSeek V4 Pro (Max) scored 83.4%; Qwen3.5 397B scored 62%. DeepSeek V4 Pro (Max) wins this benchmark. - Source ↗
Terminal-Bench 2.0
AgenticA 67.9%B 52.5%Winner: DeepSeek V4 Pro (Max)Δ 15.4Terminal-Bench 2.0: DeepSeek V4 Pro (Max) scored 67.9%; Qwen3.5 397B scored 52.5%. DeepSeek V4 Pro (Max) wins this benchmark. - Source ↗
HLE
KnowledgeA 37.7%B 28.7%Winner: DeepSeek V4 Pro (Max)Δ 9HLE: DeepSeek V4 Pro (Max) scored 37.7%; Qwen3.5 397B scored 28.7%. DeepSeek V4 Pro (Max) wins this benchmark. - Source ↗
HMMT Feb 2026
MathA 95.2%B 87.9%Winner: DeepSeek V4 Pro (Max)Δ 7.3HMMT Feb 2026: DeepSeek V4 Pro (Max) scored 95.2%; Qwen3.5 397B scored 87.9%. DeepSeek V4 Pro (Max) wins this benchmark. - Source ↗
SWE-bench Pro
CodingA 55.4%B 50.9%Winner: DeepSeek V4 Pro (Max)Δ 4.5SWE-bench Pro: DeepSeek V4 Pro (Max) scored 55.4%; Qwen3.5 397B scored 50.9%. DeepSeek V4 Pro (Max) wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | DeepSeek V4 Pro (Max) | Qwen3.5 397B | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | DeepSeek V4 Pro (Max)$0.435 input / $0.87 output | Qwen3.5 397B$0.6 input / $3.6 output | DeepSeek V4 Pro (Max) has the lower combined listed price. |
| Generation speedtokens per second | DeepSeek V4 Pro (Max)Not available | Qwen3.5 397B96 tok/s | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | DeepSeek V4 Pro (Max)Not available | Qwen3.5 397B2.44 s | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | DeepSeek V4 Pro (Max)1M | Qwen3.5 397B128K | DeepSeek V4 Pro (Max) lists the larger context window. |
Benchmark Deep Dive
AgenticDeepSeek V4 Pro (Max) wins26 benchmarks
| Benchmark | DeepSeek V4 Pro (Max) | Qwen3.5 397B | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 67.9% | 52.5% | DeepSeek V4 Pro (Max) leads |
| BrowseCompSource | 83.4% | 62% | DeepSeek V4 Pro (Max) leads |
| HLE w/ toolsSource | 48.2% | — | Not comparable |
| MCP AtlasSource | 73.6% | 46.1% | DeepSeek V4 Pro (Max) leads |
| GDPval-AASource | 1307 | 962 | DeepSeek V4 Pro (Max) leads |
| ToolathlonSource | 51.8% | 36.3% | DeepSeek V4 Pro (Max) leads |
| AA Agentic IndexSource | 36.4% | 19.9% | DeepSeek V4 Pro (Max) leads |
| APEX-Agents-AASource | 24.3% | 15.3% | DeepSeek V4 Pro (Max) leads |
| τ²-bench resultsSource | 96.2% | 95.6% | DeepSeek V4 Pro (Max) leads |
| GDPval-AASource | 40.4% | 23.1% | DeepSeek V4 Pro (Max) leads |
| AA BriefcaseSource | 932 | — | Not comparable |
| AA EnterpriseOps-GymSource | 40.4% | — | Not comparable |
| AA Harvey LABSource | 84.4% | — | Not comparable |
| AA ITBenchSource | 38.3% | — | Not comparable |
| AA Tau3 BankingSource | 25.8% | — | Not comparable |
| terminalBenchHardSource | 46.2% | — | Not comparable |
| aaTerminalBench21Source | 64% | — | Not comparable |
| Claw-EvalSource | — | 56.8% | Not comparable |
| QwenClawBenchSource | — | 51.8% | Not comparable |
| τ³-bench resultsSource | — | 68.4% | Not comparable |
| VITA-BenchSource | — | 43.7% | Not comparable |
| DeepPlanningSource | — | 37.6% | Not comparable |
| MCP-TasksSource | — | 74.2% | Not comparable |
| WideResearchSource | — | 74.0% | Not comparable |
| Gert LabsSource | — | 46.76% | Not comparable |
| ResearchClawBenchSource | — | 14.2% | Not comparable |
CodingDeepSeek V4 Pro (Max) wins9 benchmarks
| Benchmark | DeepSeek V4 Pro (Max) | Qwen3.5 397B | Result |
|---|---|---|---|
| CodeforcesSource | 3206.0 | — | Not comparable |
| SWE-bench VerifiedSource | 80.6% | 76.2% | DeepSeek V4 Pro (Max) leads |
| SWE-bench ProSource | 55.4% | 50.9% | DeepSeek V4 Pro (Max) leads |
| SWE MultilingualSource | 76.2% | — | Not comparable |
| Terminal-Bench 2.0Source | 67.9% | — | Not comparable |
| Vibe Code BenchSource | 49.93% | — | Not comparable |
| AA Coding IndexSource | 59.4% | 48.2% | DeepSeek V4 Pro (Max) leads |
| AA-SciCodeSource | 50.0% | 42.0% | DeepSeek V4 Pro (Max) leads |
| LiveCodeBench v6Source | — | 83.6% | Not comparable |
Reasoning6 benchmarks
KnowledgeDeepSeek V4 Pro (Max) wins16 benchmarks
| Benchmark | DeepSeek V4 Pro (Max) | Qwen3.5 397B | Result |
|---|---|---|---|
| MMLU-ProSource | 87.5% | 87.8% | Qwen3.5 397B leads |
| SimpleQASource | 57.9% | — | Not comparable |
| Chinese-SimpleQASource | 84.4% | — | Not comparable |
| GPQASource | 90.1% | 88.4% | DeepSeek V4 Pro (Max) leads |
| GPQA-DSource | 90.1% | — | Not comparable |
| HLESource | 37.7% | 28.7% | DeepSeek V4 Pro (Max) leads |
| Artificial Analysis Intelligence IndexSource | 44.3% | 33.7% | DeepSeek V4 Pro (Max) leads |
| AA-GPQA DiamondSource | 88.8% | 89.3% | Qwen3.5 397B leads |
| AA-HLESource | 35.9% | 27.3% | DeepSeek V4 Pro (Max) leads |
| AA-Omniscience IndexSource | -10.0% | -29.8% | DeepSeek V4 Pro (Max) leads |
| AA-Omniscience AccuracySource | 43.3% | 31.4% | DeepSeek V4 Pro (Max) leads |
| AA-Omniscience Hallucination RateSource | 94.0% | 89.1% | Qwen3.5 397B leads |
| AA Openness IndexSource | 50.0% | — | Not comparable |
| SuperGPQASource | — | 70.4% | Not comparable |
| MMLU-ReduxSource | — | 94.9% | Not comparable |
| C-EvalSource | — | 93% | Not comparable |
MathDeepSeek V4 Pro (Max) wins8 benchmarks
| Benchmark | DeepSeek V4 Pro (Max) | Qwen3.5 397B | Result |
|---|---|---|---|
| HMMT Feb 2026Source | 95.2% | 87.9% | DeepSeek V4 Pro (Max) leads |
| IMOAnswerBenchSource | 89.8% | — | Not comparable |
| ApexSource | 38.3% | — | Not comparable |
| Apex ShortlistSource | 90.2% | — | Not comparable |
| AIME26Source | — | 93.3% | Not comparable |
| HMMT Feb 2025Source | — | 94.8% | Not comparable |
| HMMT Nov 2025Source | — | 92.7% | Not comparable |
| MMAnswerBenchSource | — | 80.9% | Not comparable |
Multilingual2 benchmarks
Multimodal8 benchmarks
| Benchmark | DeepSeek V4 Pro (Max) | Qwen3.5 397B | Result |
|---|---|---|---|
| Design Arena WebsiteSource | 1264 | — | Not comparable |
| MMMU-ProSource | — | 79% | Not comparable |
| MathVisionSource | — | 88.6% | Not comparable |
| CharXivSource | — | 80.8% | Not comparable |
| VideoMMMUSource | — | 84.7% | Not comparable |
| ScreenSpot ProSource | — | 65.6% | Not comparable |
| V*Source | — | 95.8% | Not comparable |
| AA-MMMU-ProSource | — | 77.3% | Not comparable |
Frequently Asked Questions (5)
Which is better, DeepSeek V4 Pro (Max) or Qwen3.5 397B?
DeepSeek V4 Pro (Max) and Qwen3.5 397B are tied on the BenchAlign overall score, so the right pick depends on which category matters most for your use case.
Which is better for knowledge tasks, DeepSeek V4 Pro (Max) or Qwen3.5 397B?
DeepSeek V4 Pro (Max) has the edge for knowledge tasks in this comparison, averaging 60.1 versus 56.6. Inside this category, AA-Omniscience Index is the benchmark that creates the most daylight between them.
Which is better for coding, DeepSeek V4 Pro (Max) or Qwen3.5 397B?
DeepSeek V4 Pro (Max) has the edge for coding in this comparison, averaging 70.9 versus 66.5. Inside this category, AA Coding Index is the benchmark that creates the most daylight between them.
Which is better for math, DeepSeek V4 Pro (Max) or Qwen3.5 397B?
DeepSeek V4 Pro (Max) has the edge for math in this comparison, averaging 95.2 versus 90.6. Inside this category, HMMT Feb 2026 is the benchmark that creates the most daylight between them.
Which is better for agentic tasks, DeepSeek V4 Pro (Max) or Qwen3.5 397B?
DeepSeek V4 Pro (Max) has the edge for agentic tasks in this comparison, averaging 74.5 versus 56.5. Inside this category, GDPval-AA is the benchmark that creates the most daylight between them.
Related Comparisons
Explore More
Choose a model with this week’s evidence
Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.
One email each week. Unsubscribe anytime.