The Known Good Updated 8 Sep 2026 Subscribe
DeepSeek: DeepSeek V3 0324 vs OpenAI: gpt-oss-120b ·DeepSeek vs OpenAI

DeepSeek: DeepSeek V3 0324 vs OpenAI: gpt-oss-120b — head-to-head comparison

DeepSeek: DeepSeek V3 0324 profile OpenAI: gpt-oss-120b profile All comparisons
DeepSeek: DeepSeek V3 0324 — DeepSeek
Quality
#40 of 65
64based on 4 of 7 evaluations
Known Good Index
Speed
#286 of 341
24.7
Output tokens per second
Price
#163 of 450
$0.50
Blended per 1M tokens
In $0.29 · Out $1.14
Cache hit price
#188 of 344
$0.11
per 1M tokens (−62%)
Context
#347 of 487
164k
tokens (~300 pages)
OpenAI: gpt-oss-120b — OpenAI
Quality
#44 of 65
62based on 3 of 7 evaluations
Known Good Index
Speed
#10 of 341
166.1
Output tokens per second
Price
#19 of 450
$0.07
Blended per 1M tokens
In $0.04 · Out $0.17
Cache hit price
unranked
—
per 1M tokens
Context
#372 of 487
131k
tokens (~200 pages)
Summary

On the Known Good Index, DeepSeek: DeepSeek V3 0324 scores 64 and OpenAI: gpt-oss-120b scores 62 — a 2-point lead for DeepSeek: DeepSeek V3 0324. They share 2 published evaluations, with DeepSeek: DeepSeek V3 0324 ahead on 0 and OpenAI: gpt-oss-120b ahead on 2.

Blended pricing is $0.50 per 1M tokens for DeepSeek: DeepSeek V3 0324 and $0.07 for OpenAI: gpt-oss-120b — OpenAI: gpt-oss-120b is about 7.2× cheaper.

Throughput is 25 tokens per second for DeepSeek: DeepSeek V3 0324 and 166 for OpenAI: gpt-oss-120b, with time to first token at 1.5s and 0.7s. DeepSeek: DeepSeek V3 0324 accepts 164k tokens of context against 131k.

DeepSeek: DeepSeek V3 0324 is open weights; OpenAI: gpt-oss-120b is open weights under Open weights (unrestricted). 20 tracked hosts serve one or both, and 3 serve both.

Generated from tracked data on 9 Sep 2026 · how these comparisons are computed →

At a glance
MetricDeepSeek: DeepSeek V3 0324OpenAI: gpt-oss-120b
Known Good Index 64 based on 4 of 7 evaluations 62 based on 3 of 7 evaluations
Blended $/1M $0.50 $0.07
Output tok/s 25 166
Context window 164k 131k
Providers 5 18
Jump to section

Quality

Known Good Index and the capability indices, side by side

Index scores

Each index is min-max normalised across tracked models and scaled 0–100 from published evaluations. We do not run these evaluations.

MetricDeepSeek: DeepSeek V3 0324OpenAI: gpt-oss-120b
Known Good Index 64 62
Mathematics Index 57 —
Reasoning Index 65 —
Openness Index 100 60
The Known Good

Quality Breakdown

Every evaluation both models report a published score for

Evaluations both models report

Bar = DeepSeek: DeepSeek V3 0324 · marker = OpenAI: gpt-oss-120b · scores ingested from Epoch AI (CC-BY 4.0)

GPQA Diamond68 / 76
Mock AIME 2024–2538 / 89

Right-hand figures read DeepSeek: DeepSeek V3 0324 / OpenAI: gpt-oss-120b. DeepSeek: DeepSeek V3 0324 leads on 0 of 2, OpenAI: gpt-oss-120b on 2.

The Known Good

Arena Elo

Human preference rating from blind pairwise votes

Arena ratings

Elo from blind pairwise votes, ingested from Arena. Higher is better.

MetricDeepSeek: DeepSeek V3 0324OpenAI: gpt-oss-120b
Text Arena 1,375 1,366
Text Arena (style controlled) 1,396 1,352
The Known Good

Price & Cost

Published API pricing by token type

Published pricing

USD per 1M tokens. Blended is a 3:1 input:output mix. Live from OpenRouter, cross-checked against models.dev. Lower is better.

MetricDeepSeek: DeepSeek V3 0324OpenAI: gpt-oss-120b
Input $/1M $0.29 $0.04
Output $/1M $1.14 $0.17
Blended $/1M $0.50 $0.07
Cache read $/1M $0.110 —
The Known Good
We publish blended price rather than cost-per-task. Cost-per-task requires token counts from an evaluation suite we do not run. Blended price is computed from published rates and is fully reproducible — see the pricing methodology.

Context Window

Maximum input accepted and maximum output emitted

Context and output limits

Provider-declared limits, from metadata via models.dev and OpenRouter. Higher is better.

MetricDeepSeek: DeepSeek V3 0324OpenAI: gpt-oss-120b
Context window 164k 131k
Max output tokens 147k 118k
The Known Good

Speed & Latency

Mean throughput and time to first token across hosting providers, over successful measurements from the last 48 hours

Throughput and latency

Means across all tracked providers over successful measurements from the last 48 hours, derived from OpenRouter endpoint telemetry.

MetricDeepSeek: DeepSeek V3 0324OpenAI: gpt-oss-120b
Output tokens/s 25 166
Time to first token 1.5s 0.7s
The Known Good

Providers

Which hosts serve each model, and at what price and speed

20 hosts · 3 serve both
Full provider benchmarks →
Provider DeepSeek: DeepSeek V3 0324 blendedDeepSeek: DeepSeek V3 0324 tok/s OpenAI: gpt-oss-120b blendedOpenAI: gpt-oss-120b tok/s
Crusoe $0.75 36 — —
Novita $0.48 29 $0.10 61
GMICloud $0.50 20 — —
SiliconFlow $0.44 20 $0.15 10
DeepInfra $0.41 18 $0.07 105
Cerebras — — $0.45 833
Groq — — $0.26 304
Amazon Bedrock · first party — — $0.26 288
Nebius — — $0.26 281
BaseTen — — $0.20 224
Google · first party — — $0.16 196
SambaNova — — $0.34 183
Mara — — $0.30 120
Parasail — — $0.26 96
Together — — $0.26 84
Phala — — $0.26 67
Mancer 2 — — $0.17 54
AkashML — — $0.07 50
CoreWeave — — $0.07 45
DigitalOcean — — $0.14 29

Per-host pricing and telemetry from OpenRouter. A dash means that host does not serve that model.

The Known Good
Sources for this page
Pricing & providers — OpenRouter Metadata — models.dev Benchmarks — Epoch AI (CC-BY 4.0)Elo — Arena Full attribution →