Skip to content

DeepSeek V4 Pro 0813

DeepSeekOpen weightsReasoningProvisionalReleased Aug 13, 2026Input: text
Compare with Grok 4.5Source scores ↗OpenRouter ↗
Overall score
63.2
Rank 43 of 222 · 48.9–77.5
Context window
1M
1,000,000 tokens
Price per 1M tokens
$0.66 in · $1.98 out
✓ Verified on DeepSeek's pricing page
Blended price
$0.99
3:1 input:output tokens
Output speed
—
Not measured yet

Category scores

Scores per capability area with the model's rank among ranked models in that category.

  1. 1Math87.4
  2. 2Agentic & tool use#4373.5
  3. 3Knowledge#4273.1
  4. 4Reasoning60.7
  5. 5Coding#5059.4
Higher is better. #N is the model's rank in that category, where it is eligible to be ranked.

Benchmark results (33)

As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.

Agentic & tool use

Terminal-Bench 2.187.9#10 of 68
BrowseComp83.4#18 of 50
MCP Atlas73.6#21 of 42
Terminal-Bench 2.067.9#9 of 39
HLE w/ tools60.0#7 of 27
CyberGym83.3#10 of 25
AutomationBench31.8#18 of 25
Toolathlon51.8#9 of 22
Toolathlon-Verified74.1#7 of 22
Agents' Last Exam25.7#15 of 17

Coding

SWE-bench Verified80.6#14 of 98
SWE-bench Pro55.4#47 of 69
DeepSWE62.7#33 of 50
NL2Repo61.5#2 of 31
OpenHarmony Bench59.0#4 of 12
Codeforces3206.0#2 of 4
LiveCodeBench Pass@1-COT93.5#1 of 2
DSBench-FullStack71.1#1 of 2
DSBench-Hard67.2#1 of 2

Reasoning

ARC-AGI-261.3#25 of 41
ARC-AGI-190.00#21 of 36
MRCR 1M83.5#1 of 4
CorpusQA 1M62.0#1 of 2

Knowledge

GPQA90.1#25 of 117
HLE42.7#26 of 69
MMLU-Pro87.5#6 of 67
Chinese-SimpleQA84.4#1 of 2

Math

HMMT Feb 202695.2#2 of 24
IMOAnswerBench89.8#3 of 9
Apex38.3#5 of 8
Apex Shortlist90.2#1 of 4

Multilingual

SWE Multilingual76.2#23 of 46

Multimodal

Design Arena Website1258#37 of 91

Pricing by source

USD per 1M tokens.

SourceInputOutputCached in
✓ Official (DeepSeek)
checked Oct 17, 2026 · deepseek-v4-pro
$0.66$1.98$0.022
BenchLM list price$0.66$1.98$0.022
OpenRouter deepseek/deepseek-v4-pro-0813$0.66$1.98$0.022

Off-peak price; peak hours (01:00-04:00 and 06:00-10:00 UTC, weekdays) cost 2x.

BenchLM/list lists input at $1.32; BenchLM/list lists output at $3.96, which differs from the official price.

DeepSeek's pricing page, read September 25, 2026, lists `deepseek-v4-pro` (model version DeepSeek-V4-Pro-0813) with peak and off-peak rates, effective 16:00 UTC on August 16, 2026 per DeepSeek's August 13, 2026 V4-Pro GA announcement and change log (https://api-docs.deepseek.com/news/news260813, https://api-docs.deepseek.com/updates/). The stored values are the peak rates: $1.32 cache-miss input / $0.044 cache-hit input / $3.96 output per million tokens. Off-peak rates are half of peak: $0.66 / $0.022 / $1.98. Peak hours are 01:00-04:00 and 06:00-10:00 UTC, Monday through Friday, excluding Chinese public holidays; all other hours are off-peak. The page lists a 1M context length, a 384K maximum output, and a 500-request concurrency limit, and no Batch API discount. DeepSeek's September 10, 2026 V4.1 Flash announcement said `deepseek-v4-pro` requests would route to DeepSeek-V4.1-Flash and bill at the V4.1 Flash rate from 04:00 UTC on September 14, 2026 until V4.1 Pro launches. DeepSeek then said V4 Pro API service continues after September 14, 2026 with the billing method unchanged and that it will give notice of any change: on its pricing page, read September 17, 2026, and in its change log entry for September 10, read September 25, 2026 (https://api-docs.deepseek.com/news/news260910). This row previously stored the flat $0.435 cache-miss input / $0.003625 cache-hit input / $0.87 output rate from the pricing page checked August 12, 2026, before the peak/off-peak schedule took effect.

Price history

  • Oct 17, 2026: $1.32 / $3.96
  • Oct 17, 2026: $0.66 / $1.98

Open weights

From the Hugging Face repository.

Parameters
1.65T
Licence
mit
Downloads (30 days)
156.7K

Speed

Measured by us through OpenRouter (median of trials).

Not measured yet.

Related models

Similar score