Skip to main content

Model comparison

DeepSeek V3 vs Soofi S 30B-A3B

Data verified

Head-to-head evidence from 2 shared benchmark results across 1 category. Overall scores shown here use BenchLM's provisional ranking lane.

DeepSeek
38/100
Margin
7.0pts
winning →
Soofi Project
45/100
1 category wins0 category wins

Evidence parity. DeepSeek V3 and Soofi S 30B-A3B share 2 comparable benchmark results. 1 of 8 categories are comparable. 21 results are unique to DeepSeek V3; 6 to Soofi S 30B-A3B.

Updated July 15, 2026
Shared results
2
DeepSeek V3 only
21
Soofi S 30B-A3B only
6
Comparable categories
1 / 8

Pick Soofi S 30B-A3B if you want the stronger benchmark profile. DeepSeek V3 only becomes the better choice if knowledge is the priority.

Confidence note. This is a partial-evidence comparison with 2 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

Soofi S 30B-A3B is clearly ahead on the provisional aggregate, 45 to 38. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

DeepSeek V3 is also the more expensive model on tokens at $0.27 input / $1.10 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. Soofi S 30B-A3B gives you the larger context window at 1M, compared with 128K for DeepSeek V3.

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 scores and score margins for DeepSeek V3 and Soofi S 30B-A3B
CategoryDeepSeek V3ΔSoofi S 30B-A3B
KnowledgeDeepSeek V372.8Margin 22.9Soofi S 30B-A3B49.9
CodingDeepSeek V339.2MarginNo overlapSoofi S 30B-A3BNot measured
MathDeepSeek V31.7MarginNo overlapSoofi S 30B-A3BNot measured
Inst. FollowingDeepSeek V386.1MarginNo overlapSoofi S 30B-A3BNot measured

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

More
A · DeepSeek V3B · Soofi S 30B-A3B
  1. MMLU-Pro

    Knowledge
    Source ↗
    A 75.9%B 51.4%
    Winner: DeepSeek V3Δ 24.5
    MMLU-Pro: DeepSeek V3 scored 75.9%; Soofi S 30B-A3B scored 51.4%. DeepSeek V3 wins this benchmark.
  2. GPQA

    Knowledge
    Source ↗
    A 59.1%B 43.4%
    Winner: DeepSeek V3Δ 15.7
    GPQA: DeepSeek V3 scored 59.1%; Soofi S 30B-A3B scored 43.4%. DeepSeek V3 wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

MetricDeepSeek V3Soofi S 30B-A3BComparison
Input / output priceUSD per 1M tokensDeepSeek V3$0.27 input / $1.1 outputSoofi S 30B-A3B$0 input / $0 outputSoofi S 30B-A3B has the lower combined listed price.
Generation speedtokens per secondDeepSeek V3Not availableSoofi S 30B-A3BNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenDeepSeek V3Not availableSoofi S 30B-A3BNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensDeepSeek V3128KSoofi S 30B-A3B1MSoofi S 30B-A3B lists the larger context window.

Benchmark Deep Dive

Agentic
BenchmarkDeepSeek V3Soofi S 30B-A3BResult
AA Agentic IndexSource 1.6%Not comparable
τ²-bench resultsSource 22.8%Not comparable
GDPval-AASource 0.0%Not comparable
GDPval-AASource 217Not comparable
Coding
BenchmarkDeepSeek V3Soofi S 30B-A3BResult
LiveCodeBenchSource 37.6%Not comparable
SWE-bench VerifiedSource 42%Not comparable
AA Coding IndexSource 23.0%Not comparable
Terminal-Bench HardSource 6.8%Not comparable
AA-SciCodeSource 35.4%Not comparable
HumanEvalSource 73.8%Not comparable
Reasoning
BenchmarkDeepSeek V3Soofi S 30B-A3BResult
AA-LCRSource 29.0%Not comparable
CritPtSource 0.0%Not comparable
BBHSource 78.8%Not comparable
DROPSource 66.5%Not comparable
KnowledgeDeepSeek V3 wins
BenchmarkDeepSeek V3Soofi S 30B-A3BResult
GPQASource 59.1%43.4%DeepSeek V3 leads
MMLU-ProSource 75.9%51.4%DeepSeek V3 leads
Artificial Analysis Intelligence IndexSource 14.2%Not comparable
AA-GPQA DiamondSource 55.7%Not comparable
AA-HLESource 3.6%Not comparable
AA-Omniscience IndexSource -41.3%Not comparable
AA-Omniscience AccuracySource 25.4%Not comparable
AA-Omniscience Hallucination RateSource 89.4%Not comparable
GPQA-DSource 43.4%Not comparable
AGIEvalSource 66.9%Not comparable
Math
BenchmarkDeepSeek V3Soofi S 30B-A3BResult
FrontierMath v2 (Tiers 1-3)Source 1.724%Not comparable
GSM8KSource 86.1%Not comparable
Multimodal
BenchmarkDeepSeek V3Soofi S 30B-A3BResult
Design Arena WebsiteSource 1154Not comparable
Inst. Following
BenchmarkDeepSeek V3Soofi S 30B-A3BResult
IFEvalSource 86.1%Not comparable
AA-IFBenchSource 34.8%Not comparable
Frequently Asked Questions (2)

Which is better, DeepSeek V3 or Soofi S 30B-A3B?

Soofi S 30B-A3B is ahead on BenchLM's provisional leaderboard, 45 to 38. The biggest single separator in this matchup is MMLU-Pro, where the scores are 75.9% and 51.4%.

Which is better for knowledge tasks, DeepSeek V3 or Soofi S 30B-A3B?

DeepSeek V3 has the edge for knowledge tasks in this comparison, averaging 72.8 versus 49.9. Inside this category, MMLU-Pro is the benchmark that creates the most daylight between them.

Self-host vs API cost

Estimates at 50,000 req/day · 1000 tokens/req average.

DeepSeek V3
API / mo$1,028
Self-host / mo$18,221
Break-even1.2B/day
Soofi S 30B-A3B
API / mo$0
Self-host / moNot listed
Break-even
Proprietary model — self-hosting not applicable.
Model the full break-even

Related Comparisons

Last updated: July 15, 2026

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.