Model comparison
DeepSeek V4 Flash (Max) vs Soofi S 30B-A3B
Head-to-head evidence from 3 shared benchmark results across 1 category. Overall scores shown here use BenchLM's provisional ranking lane.
Verified leaderboard positions: DeepSeek V4 Flash (Max) #20; Soofi S 30B-A3B unranked
Evidence parity. DeepSeek V4 Flash (Max) and Soofi S 30B-A3B share 3 comparable benchmark results. 1 of 8 categories are comparable. 37 results are unique to DeepSeek V4 Flash (Max); 5 to Soofi S 30B-A3B.
Updated July 15, 2026- Shared results
- 3
- DeepSeek V4 Flash (Max) only
- 37
- Soofi S 30B-A3B only
- 5
- Comparable categories
- 1 / 8
Pick DeepSeek V4 Flash (Max) if you want the stronger benchmark profile. Soofi S 30B-A3B only becomes the better choice if you want the cheaper token bill or you would rather avoid the extra latency and token burn of a reasoning model.
Confidence note. This is a partial-evidence comparison with 3 shared benchmark results across 1 evidence category; 1 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 Flash (Max) is clearly ahead on the provisional aggregate, 70 to 45. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
DeepSeek V4 Flash (Max)'s sharpest advantage is in knowledge, where it averages 55.6 against 49.9. The single biggest benchmark swing on the page is GPQA, 88.1% to 43.4%.
DeepSeek V4 Flash (Max) is also the more expensive model on tokens at $0.14 input / $0.28 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Soofi S 30B-A3B. That is roughly Infinityx on output cost alone. DeepSeek V4 Flash (Max) is the reasoning model in the pair, while Soofi S 30B-A3B 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.
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 Flash (Max) | Δ | Soofi S 30B-A3B |
|---|---|---|---|
| Knowledge | DeepSeek V4 Flash (Max)55.6 | Margin← 5.7 | Soofi S 30B-A3B49.9 |
| Agentic | DeepSeek V4 Flash (Max)63.8 | MarginNo overlap | Soofi S 30B-A3BNot measured |
| Coding | DeepSeek V4 Flash (Max)62.4 | MarginNo overlap | Soofi S 30B-A3BNot measured |
| Math | DeepSeek V4 Flash (Max)94.8 | MarginNo overlap | Soofi S 30B-A3BNot measured |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
GPQA
KnowledgeA 88.1%B 43.4%Winner: DeepSeek V4 Flash (Max)Δ 44.7GPQA: DeepSeek V4 Flash (Max) scored 88.1%; Soofi S 30B-A3B scored 43.4%. DeepSeek V4 Flash (Max) wins this benchmark. - Source ↗
MMLU-Pro
KnowledgeA 86.2%B 51.4%Winner: DeepSeek V4 Flash (Max)Δ 34.8MMLU-Pro: DeepSeek V4 Flash (Max) scored 86.2%; Soofi S 30B-A3B scored 51.4%. DeepSeek V4 Flash (Max) wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | DeepSeek V4 Flash (Max) | Soofi S 30B-A3B | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | DeepSeek V4 Flash (Max)$0.14 input / $0.28 output | Soofi S 30B-A3B$0 input / $0 output | Soofi S 30B-A3B has the lower combined listed price. |
| Generation speedtokens per second | DeepSeek V4 Flash (Max)Not available | Soofi S 30B-A3BNot available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | DeepSeek V4 Flash (Max)Not available | Soofi S 30B-A3BNot available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | DeepSeek V4 Flash (Max)1M | Soofi S 30B-A3B1M | Listed context windows are equal. |
Benchmark Deep Dive
Agentic9 benchmarks
| Benchmark | DeepSeek V4 Flash (Max) | Soofi S 30B-A3B | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 56.9% | — | Not comparable |
| BrowseCompSource | 73.2% | — | Not comparable |
| HLE w/ toolsSource | 45.1% | — | Not comparable |
| MCP AtlasSource | 69% | — | Not comparable |
| GDPval-AASource | 1189 | — | Not comparable |
| ToolathlonSource | 47.8% | — | Not comparable |
| AA Agentic IndexSource | 31.1% | — | Not comparable |
| τ²-bench resultsSource | 95% | — | Not comparable |
| GDPval-AASource | 34.4% | — | Not comparable |
Coding10 benchmarks
| Benchmark | DeepSeek V4 Flash (Max) | Soofi S 30B-A3B | Result |
|---|---|---|---|
| LiveCodeBench Pass@1-COTSource | 91.6% | — | Not comparable |
| CodeforcesSource | 3052.0 | — | Not comparable |
| SWE-bench VerifiedSource | 79% | — | Not comparable |
| SWE-bench ProSource | 52.6% | — | Not comparable |
| SWE MultilingualSource | 73.3% | — | Not comparable |
| Terminal-Bench 2.0Source | 56.9% | — | Not comparable |
| AA Coding IndexSource | 56.2% | — | Not comparable |
| Terminal-Bench HardSource | 35.6% | — | Not comparable |
| AA-SciCodeSource | 44.9% | — | Not comparable |
| HumanEvalSource | — | 73.8% | Not comparable |
Reasoning6 benchmarks
KnowledgeDeepSeek V4 Flash (Max) wins13 benchmarks
| Benchmark | DeepSeek V4 Flash (Max) | Soofi S 30B-A3B | Result |
|---|---|---|---|
| MMLU-ProSource | 86.2% | 51.4% | DeepSeek V4 Flash (Max) leads |
| SimpleQASource | 34.1% | — | Not comparable |
| Chinese-SimpleQASource | 78.9% | — | Not comparable |
| GPQASource | 88.1% | 43.4% | DeepSeek V4 Flash (Max) leads |
| GPQA-DSource | 88.1% | 43.4% | DeepSeek V4 Flash (Max) leads |
| HLESource | 34.8% | — | Not comparable |
| Artificial Analysis Intelligence IndexSource | 40.3% | — | Not comparable |
| AA-GPQA DiamondSource | 89.4% | — | Not comparable |
| AA-HLESource | 32.1% | — | Not comparable |
| AA-Omniscience IndexSource | -22.9% | — | Not comparable |
| AA-Omniscience AccuracySource | 37.2% | — | Not comparable |
| AA-Omniscience Hallucination RateSource | 95.8% | — | Not comparable |
| AGIEvalSource | — | 66.9% | Not comparable |
Math5 benchmarks
Multimodal1 benchmarks
| Benchmark | DeepSeek V4 Flash (Max) | Soofi S 30B-A3B | Result |
|---|---|---|---|
| Design Arena WebsiteSource | 1242 | — | Not comparable |
Inst. Following1 benchmarks
| Benchmark | DeepSeek V4 Flash (Max) | Soofi S 30B-A3B | Result |
|---|---|---|---|
| AA-IFBenchSource | 79.2% | — | Not comparable |
Frequently Asked Questions (2)
Which is better, DeepSeek V4 Flash (Max) or Soofi S 30B-A3B?
DeepSeek V4 Flash (Max) is ahead on BenchLM's provisional leaderboard, 70 to 45. The biggest single separator in this matchup is GPQA, where the scores are 88.1% and 43.4%.
Which is better for knowledge tasks, DeepSeek V4 Flash (Max) or Soofi S 30B-A3B?
DeepSeek V4 Flash (Max) has the edge for knowledge tasks in this comparison, averaging 55.6 versus 49.9. Inside this category, GPQA is the benchmark that creates the most daylight between them.
Related Comparisons
Explore More
The AI models change fast. We track them for you.
A weekly brief for engineers and researchers covering new models, ranking shifts, and pricing changes.
Free. No spam. Unsubscribe anytime.