Skip to main content

Model comparison

DeepSeek V3.1 vs GPT-OSS 20B

Data verified

Head-to-head evidence from 12 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

53.64/100
Margin
10.9pts
← winning
42.74/100
0 category wins0 category wins

Public leaderboard positions: DeepSeek V3.1 #95 (Supported); GPT-OSS 20B #159 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. DeepSeek V3.1 and GPT-OSS 20B share 12 comparable benchmark results. 0 of 8 categories are comparable. 0 results are unique to DeepSeek V3.1; 6 to GPT-OSS 20B.

Updated July 22, 2026
Shared results
12
DeepSeek V3.1 only
0
GPT-OSS 20B only
6
Comparable categories
0 / 8

Benchmark data for DeepSeek V3.1 and GPT-OSS 20B is coming soon on BenchLM.

Confidence note. This is a partial-evidence comparison with 12 shared benchmark results across 6 evidence categories; 0 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

BenchLM has partial data for these models, but not enough overlapping benchmark coverage to produce a fair score-level comparison yet.

Category breakdown

Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.

Operational comparison

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

MetricDeepSeek V3.1GPT-OSS 20BComparison
Input / output priceUSD per 1M tokensDeepSeek V3.1$0 input / $0 outputGPT-OSS 20B$0 input / $0 outputListed prices are equal.
Generation speedtokens per secondDeepSeek V3.1Not availableGPT-OSS 20B313 tok/sA complete speed comparison is not available.
First-answer latencyseconds to first tokenDeepSeek V3.1Not availableGPT-OSS 20B0.65 sA complete latency comparison is not available.
Context windowmaximum listed tokensDeepSeek V3.1128KGPT-OSS 20B128KListed context windows are equal.

Benchmark Deep Dive

Agentic
BenchmarkDeepSeek V3.1GPT-OSS 20BResult
τ²-bench resultsSource 34.8%60.2%GPT-OSS 20B leads
AA Agentic IndexSource 3.1%Not comparable
APEX-Agents-AASource 0.7%Not comparable
GDPval-AASource 3.0%Not comparable
GDPval-AASource 559Not comparable
Coding
BenchmarkDeepSeek V3.1GPT-OSS 20BResult
AA-SciCodeSource 36.7%34.4%DeepSeek V3.1 leads
React Native EvalsSource 71%Not comparable
AA Coding IndexSource 20.7%Not comparable
Reasoning
BenchmarkDeepSeek V3.1GPT-OSS 20BResult
AA-LCRSource 45.0%30.7%DeepSeek V3.1 leads
CritPtSource 0.0%1.4%GPT-OSS 20B leads
Knowledge
BenchmarkDeepSeek V3.1GPT-OSS 20BResult
Artificial Analysis Intelligence IndexSource 21.1%14.9%DeepSeek V3.1 leads
AA-GPQA DiamondSource 73.5%68.8%DeepSeek V3.1 leads
AA-HLESource 6.3%9.8%GPT-OSS 20B leads
AA-Omniscience IndexSource -41.1%-63.9%DeepSeek V3.1 leads
AA-Omniscience AccuracySource 23.1%15.5%DeepSeek V3.1 leads
AA-Omniscience Hallucination RateSource 83.5%94.1%DeepSeek V3.1 leads
Multimodal
BenchmarkDeepSeek V3.1GPT-OSS 20BResult
Design Arena WebsiteSource 1152882DeepSeek V3.1 leads
Inst. Following
BenchmarkDeepSeek V3.1GPT-OSS 20BResult
AA-IFBenchSource 37.8%65.1%GPT-OSS 20B leads
Frequently Asked Questions (3)

Can I compare DeepSeek V3.1 and GPT-OSS 20B on BenchLM yet?

Not fully yet. BenchLM is tracking both models, but the sourced benchmark breakdown for this comparison is still coming soon.

Why does this comparison show “coming soon”?

BenchLM only shows category winners and benchmark-level calls when we have sourced results that can be compared fairly. For these models, the public benchmark coverage is not complete enough yet.

What data is available for DeepSeek V3.1 and GPT-OSS 20B today?

DeepSeek V3.1: $0.00 input / $0.00 output per 1M tokens GPT-OSS 20B: $0.00 input / $0.00 output per 1M tokens Both model pages still include creator, context window, reasoning mode, and other metadata while benchmark coverage fills in.

Related Comparisons

Last updated: July 22, 2026

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.