The Known Good Updated 4 Sep 2026 Subscribe
DeepSeek ·Open weights ·Released March 2025·Tracked since 25 Jul 2026

DeepSeek: DeepSeek V3 0324 — quality, speed & price analysis

Compare with another model Provider benchmarks Data sources
Quality
#40 of 65
64based on 4 of 7 evaluations
Known Good Index
Speed
#295 of 338
23.9
Output tokens per second
Price
#143 of 442
$0.44
Blended per 1M tokens
In $0.25 · Out $1.00
Cache hit price
unranked
per 1M tokens
Context
#334 of 476
164k
tokens (~300 pages)
Summary

DeepSeek: DeepSeek V3 0324 is in the third quarter of the models we track for quality, and inexpensive for its standing, and notably slow. It accepts text input, outputs text, and supports a 164k-token context window.

It scores 64 on the Known Good Index, in line with the tracked average of 65, from 4 of the seven component evaluations. Measured against the tracked average, its strongest result is MATH Level 5 (76 against 54) and its weakest is Mock AIME 2024–25 (38 against 54).

Pricing is $0.25 per 1M input tokens (below the $2.07 average) and $1.00 per 1M output (below the $9.51 average). At 24 tokens per second it sits in the slowest quarter of the models we track.

Generated from tracked data on 4 Sep 2026 · how these comparisons are computed →

Technical specifications
ReasoningNo
Input modalitytext
Output modalitytext
Context window163,840 tokens
Max output147,456 tokens
WeightsOpen weights
Knowledge cutoffJuly 2024
Tool callingYes
Providers5 hosts · compare →
Highlights — where DeepSeek: DeepSeek V3 0324 sits
Jump to section
Models shown

Quality

How DeepSeek: DeepSeek V3 0324 compares against every tracked model

Known Good IndexCodingMathematicsReasoning
Known Good Index

Composite of GPQA Diamond, Mock AIME 2024–25, MATH Level 5, SWE-bench Verified, LiveBench, Humanity's Last Exam and Terminal-Bench, normalised 0–100. Scores are ingested from Epoch AI, LiveBench; we do not run these evaluations.

💡 A lightbulb marks a model its provider publishes as a reasoning model
The Known Good

Quality Breakdown

DeepSeek: DeepSeek V3 0324 on each evaluation feeding the index, against the tracked average

Evaluation scores

Bar = DeepSeek: DeepSeek V3 0324 · grey marker = average across tracked models · scores ingested from Epoch AI, LiveBench

GPQA Diamond68 / 64
Mock AIME 2024–2538 / 54
MATH Level 576 / 54
LiveBench67 / 52
The Known Good

Arena Elo

Human preference rating from blind pairwise votes

Text Arena

Elo from blind pairwise votes, ingested from Arena

1375
Elo · 95% CI ±4
#161
of 406 ranked models
45k
votes counted
1396
style-controlled (#149)

Ingested 4 Sep 2026.

The Known Good

Price & Cost

Published API pricing by token type

BlendedInput vs outputQuality per dollar
Blended Price

USD per 1M tokens at a 3:1 input:output blend. Live from OpenRouter, cross-checked against models.dev. Lower is better. Ranked against the 442 models priced above zero, which is the population the chart draws; excludes 29 published at no cost (free and promotional tiers).

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 tokens accepted

Context windowMax output tokens
Context Window

Maximum input context from provider metadata via models.dev and OpenRouter. Higher is better.

The Known Good

Speed

Mean output throughput across hosting providers, over successful measurements from the last 48 hours

Output speedBy provider

Latency

Time to first token, mean across providers over successful measurements from the last 48 hours

Latency (TTFT)

Seconds to first streamed chunk, mean across tracked providers over successful measurements from the last 48 hours. Derived from OpenRouter endpoint telemetry. Lower is better.

The Known Good

Providers

Hosts serving DeepSeek: DeepSeek V3 0324, with their own price and speed

5 providers
Full provider benchmarks →
ProviderInput $/MOutput $/M BlendedSpeedTTFTContext
Crusoe $0.50 $1.50 $0.75 32 0.4s 164k
Novita $0.27 $1.12 $0.48 29 1.1s 164k
GMICloud $0.29 $1.14 $0.50 23 3.3s 164k
SiliconFlow $0.25 $1.00 $0.44 20 1.5s 164k
DeepInfra $0.24 $0.90 $0.41 6 5.1s 164k

Per-host pricing and telemetry from OpenRouter. Speed is output tokens per second; TTFT is time to first token.

The Known Good

History

How price and quality have moved since we began tracking this model

Quality & price over time

Since 25 Jul 2026

Known Good Index 64 · based on 4 of 7 evaluations
Blended price $0.44 per 1M · observed 4 Sep 2026
Output speed 24 tok/s · observed 4 Sep 2026
Time to first token 1.7s
19 Aug 2026 · price change
DeepSeek: DeepSeek V3 0324 price down 9%
Blended 3:1 price moved from $0.48 to $0.44 per 1M tokens (OpenRouter).
The Known Good

Comparisons

Head-to-head pages pairing DeepSeek: DeepSeek V3 0324 with other tracked models

Sources for this page
Pricing & providers — OpenRouter Metadata — models.dev Benchmarks — Epoch AI (CC-BY 4.0)Benchmarks — LiveBench (Apache-2.0)Elo — Arena Full attribution →