DeepSeek Open weights: deepseek-ai/DeepSeek-V4-ProReasoning modelCompare with another model

DeepSeek V4 Pro

Reasoning effort

DeepSeek V4 Pro is a DeepSeek open-weights reasoning model, released 13 Aug 2026. Its best configuration (max reasoning effort) ranks #80 of 775 on the BenchLeader Index at 62.7 ±3.3, in the upper half. The ± is the point: 159 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in instruction following (74) and lowest in maths (56). At $1.98 per million tokens blended it is pricier than most ranked models. Output speed of 103 tokens per second puts it faster than most, with a first answer in 47.4 s. It has been measured at 6 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 14 Oct 2026.

Blended price: $1.98/M; $1.32 in · $3.96 out
Output speed: 103 tok/s; measured by Artificial Analysis
First answer: 47 s; first token 1.51 s
Context: 1.0M
Full answer: 53 s; median, reasoning included
Cost per run: $0.674; one full Intelligence Index run
Released: 13 Aug 2026

How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.

Overall index63
Reasoning65
Coding58
Agents & tools64
Maths56
Knowledge60
Instruction following74
Long context65
Composite70

Reasoning-effort configurations

The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.

Effort	Index	Rank	Speed	First answer	Chat reply cost	Categories
no reasoning	48.9	#406	102 tok/s	1.75 s	$0.0017	Agents & tools 65 · Coding 34 · Composite 52 · Instruction following 48 · Knowledge 56 · Long context 50 · Maths 44 · Reasoning 40
low	–	–	41 tok/s	1.51 s	$0.0009	Reasoning 60
high	58.1	#163	101 tok/s	21 s	$0.0004	Agents & tools 57 · Coding 55 · Composite 63 · Instruction following 70 · Knowledge 57 · Long context 60 · Reasoning 59
xhigh	–	–	41 tok/s	1.51 s	$0.0009
maxbest	62.7	#80	103 tok/s	47 s	$0.0017	Agents & tools 64 · Coding 58 · Composite 70 · Instruction following 74 · Knowledge 60 · Long context 65 · Maths 56 · Reasoning 65
not stated	53.6	#278	41 tok/s	1.51 s	$0.0009	Agents & tools 44 · Coding 54 · Human preference 58 · Maths 60 · Reasoning 57

How its index has moved

8 Sept 2026 to 21 Oct 2026

max
high
not stated
no reasoning

The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.

Benchmark results

See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Metadata

Benchmark	max	Source	Trend
AA Openness Indexnot in index	44.4#27	Artificial Analysis

Reasoning

Benchmark	no reasoning	low	high	max	not stated	Source
GPQA Diamond	73.2%#157	–	–	91.7%#29	–	Epoch AI Benchmarking Hub
Arena Hard Promptseffort not stated	–	–	–	–	1480#65	Arena (LMArena)
LiveBench 1.0 Reasoningnot in indexeffort not stated	–	–	–	–	85.8%#34	LiveBench
GPQA Diamond (AA)not in index	71.7%#281	–	90.5%#59	92.8%#28	–	Artificial Analysis
Humanity's Last Exam (AA)not in index	10.6%#310	–	35.2%#116	41.0%#79	–	Artificial Analysis
GPQA Diamond (Vals)not in index	–	–	–	92.4%#16	–	Vals AI
Kagi LLM Benchmarkeffort not stated	–	–	–	–	53.5%#70	Kagi LLM Benchmark
ARC-AGI-1	13.0%#217	90.5%#68	87.2%#89	90.0%#74	–	ARC Prize
ARC-AGI-2	0.8%#217	56.3%#92	59.7%#84	61.3%#77	–	ARC Prize
CritPt	0.0%#310	–	10.0%#119	18.0%#70	–	Artificial Analysis
Chess Puzzlesnot in index	4.0%#132	–	–	46.0%#26	–	Epoch AI Benchmarking Hub
Mystery Game Puzzlesnot in index	17.0%#74	–	–	43.0%#15	–	Epoch AI Benchmarking Hub
LMCAnot in index	–	–	–	41.2%#105	45.5%#88	Epoch AI Benchmarking Hub
DTBenchnot in index	–	–	–	90.7%#79	93.9%#60	Epoch AI Benchmarking Hub
ForecastBenchnot in indexeffort not stated	–	–	–	–	60.9%#70	ForecastBench

Coding

Benchmark	no reasoning	high	max	not stated	Source
SWE-bench Verified (Epoch)	–	–	77.6%#6	–	Epoch AI Benchmarking Hub
SciCode	39.9%#166	–	51.0%#100	–	SciCode
WeirdML	–	–	66.2%#39	–	WeirdML
FrontierCode	17.6%#40	28.6%#33	–	–	Cognition
Arena Codingeffort not stated	–	–	–	1501#77	Arena (LMArena)
Arena WebDev	–	1582#33	–	1447#70	Arena (LMArena)
LiveBench 1.0 Codingnot in indexeffort not stated	–	–	–	77.2%#39	LiveBench
SciCode (AA)not in index	39.9%#174	–	51.0%#99	–	Artificial Analysis
LiveCodeBench	–	–	87.5%#13	–	Vals AI
IOI	–	–	51.6%#24	–	Vals AI
SWE-bench (Vals)not in index	–	–	96.4%#2	–	Vals AI
Code Migrationnot in index	–	–	41.5%#28	–	Vals AI
ProgramBenchnot in index	–	–	0.0%#32	–	Vals AI
Vibe Code Bench 1-100not in index	–	–	17.5%#15	–	Vals AI
Vibe Code Bench v1.1not in index	–	–	82.3%#22	–	Vals AI
SWE Atlas: Codebase QnAnot in indexeffort not stated	–	–	–	27.1%#16	Scale AI SEAL
SWE Atlas: Test Writingnot in indexeffort not stated	–	–	–	27.1%#20	Scale AI SEAL
ALE-Benchnot in index	521.7#103	–	1403.2#27	–	Epoch AI Benchmarking Hub
Terminal-Bench 4.0 (AA)not in index	1.5%#119	–	14.1%#72	–	Artificial Analysis
Terminal-Bench 2.1 (AA)not in index	–	64.8%#85	78.7%#49	–	Artificial Analysis
Startrise Frontendnot in indexeffort not stated	–	–	–	62.7%#28	Startrise
FrontierCode 1.1 (AA)not in index	–	–	38.8%#28	–	Artificial Analysis

Agents & tools

Benchmark	no reasoning	high	max	not stated	Source
APEX-Agentseffort not stated	–	–	–	47.3%#33	Mercor
Arena Agent	–	0#29	–	–	Arena (LMArena)
LiveBench Agentic Codingnot in indexeffort not stated	–	–	–	55.0%#30	LiveBench
Terminal-Bench Hard	36.4%#71	41.7%#50	46.2%#30	–	Artificial Analysis
τ²-Bench Telecom (AA)not in index	91.2%#58	94.2%#32	96.2%#14	–	Artificial Analysis
Terminal-Bench 2.1 (Vals)	–	–	54.7%#49	50.2%#57	Vals AI
HiL-Bencheffort not stated	–	–	–	38.5%#18	Scale AI SEAL
GDPval-AA v2.1not in index	27.9%#149	32.6%#129	47.4%#63	–	Artificial Analysis
τ³-Banking (AA)	–	26.2%#76	39.6%#31	–	Artificial Analysis
ITBench SRE (AA)not in index	–	–	38.3%#22	–	Artificial Analysis
Analyst Agent (AA)not in index	–	–	18.8%#37	–	Artificial Analysis
APEX-Agents (AA)not in index	–	–	24.3%#19	–	Artificial Analysis
CyberBenchnot in index	–	–	72.9%#14	–	Vals AI
Finance Agent v2not in index	–	–	50.4%#43	–	Vals AI
Harvey's Legal Agent Benchmarknot in index	–	–	10.8%#12	–	Vals AI
Legal Research Benchnot in index	–	–	40.9%#27	–	Vals AI
SkillsBenchnot in index	–	–	53.8%#16	–	Vals AI
SREBenchnot in index	–	–	1.9%#24	–	Vals AI
Tax Agent Benchnot in index	–	–	26.8%#33	–	Vals AI
Terminal-Bench 2.0 (Vals)not in indexeffort not stated	–	–	–	56.2%#14	Vals AI
Terminal-Bench 4.0 (Vals)not in index	–	–	14.1%#33	–	Vals AI
Terminal-Bench Sciencenot in index	–	–	4.3%#25	–	Vals AI
Vending-Bench 2not in indexeffort not stated	–	–	–	3284.5#48	Andon Labs
AutomationBenchnot in index	34.9%#108	–	56.7%#51	–	Artificial Analysis
GDP.pdf (AA)not in index	3.6%#174	–	11.4%#116	–	Artificial Analysis
EnterpriseOps-Gymnot in index	–	–	49.6%#4	–	Artificial Analysis
AA-Briefcase v1.1not in index	–	–	1256#46	–	Artificial Analysis
Terminal-Bench-Science 0.1 (AA)not in index	–	–	5.7%#35	–	Artificial Analysis

Maths

Benchmark	no reasoning	max	not stated	Source
FrontierMath Tiers 1–3	–	64.6%#36	–	Epoch AI Benchmarking Hub
FrontierMath Tier 4	–	26.8%#42	–	Epoch AI Benchmarking Hub
OTIS Mock AIME	46.7%#186	98.6%#22	–	Epoch AI Benchmarking Hub
ProofBench	–	16.0%#56	50.0%#30	Vals AI
LiveBench Mathematicsnot in indexeffort not stated	–	–	95.1%#17	LiveBench
AIME 2026possible contamination	–	96.7%#6	–	MathArena
HMMT February 2026possible contamination	–	93.9%#10	–	MathArena
USAMO 2026not in indexpossible contamination	–	60.7%#4	–	MathArena
MathArena Apexpossible contamination	–	28.1%#10	–	MathArena
Arena Mathseffort not stated	–	–	1445#80	Arena (LMArena)

Knowledge

Benchmark	no reasoning	high	xhigh	max	not stated	Source
SimpleQA Verified	–	–	–	52.9%#23	–	Epoch AI Benchmarking Hub
LiveBench 1.0 Data Analysisnot in indexeffort not stated	–	–	–	–	79.2%#18	LiveBench
AA-Omniscience	-12.4#209	-10.6#196	–	0.8#128	–	Artificial Analysis
MMLU-Pro	–	–	–	87.3%#32	–	Vals AI
LegalBench	–	–	–	82.4%#71	–	Vals AI
CorpFin	–	–	–	65.4%#28	–	Vals AI
TaxEval	–	–	–	73.1%#52	–	Vals AI
CaseLaw v2not in index	–	–	–	59.4%#27	–	Vals AI
Excel Modeling Benchmarknot in index	–	–	–	52.8%#51	–	Vals AI
MedCodenot in index	–	–	–	42.5%#49	–	Vals AI
MedScribenot in index	–	–	–	80.2%#57	–	Vals AI
Public Benefits Benchnot in index	–	–	–	62.9%#25	–	Vals AI
DrugDiscoveryBenchnot in index	–	–	31.7%#13	–	–	Scale AI SEAL
AA-Omniscience: accuracynot in index	41.5%#111	41.4%#112	–	49.1%#66	–	Artificial Analysis
AA-Omniscience: non-hallucinationnot in index	8.1%#467	11.3%#388	–	5.2%#517	–	Artificial Analysis

Instruction following

Benchmark	no reasoning	high	max	not stated	Source
LiveBench Languagenot in indexeffort not stated	–	–	–	82.1%#26	LiveBench
LiveBench Instruction Followingnot in indexeffort not stated	–	–	–	67.7%#35	LiveBench
IFBench	45.8%#198	71.3%#57	76.5%#20	–	Artificial Analysis
Arena Instruction Followingnot in indexeffort not stated	–	–	–	1453#60	Arena (LMArena)

Human preference

Benchmark	not stated	Source
Arena Text	1458#64	Arena (LMArena)
EQ-Bench 4	1166#17	EQ-Bench
Arena Creative Writingnot in index	1446#47	Arena (LMArena)
Arena Multi-turnnot in index	1475#49	Arena (LMArena)
Arena Longer Queriesnot in index	1472#59	Arena (LMArena)

Long context

Benchmark	no reasoning	high	max	Source	Trend
AA-LCR	50.7%#316	70.3%#208	80.3%#70	Artificial Analysis
MLCRnot in index	–	–	17.8%#36	Artificial Analysis

Composite

Benchmark	no reasoning	high	max	not stated	Source
Epoch Capabilities Indexnot in indexeffort not stated	–	–	–	155.6#34	Epoch AI Benchmarking Hub
LiveBench 1.0not in indexeffort not stated	–	–	–	77.4%#23	LiveBench
AA Intelligence Index v4.3.2	20.4#228	30.1#120	36#85	–	Artificial Analysis
Vals Indexnot in index	–	–	53.0#16	–	Vals AI

Where to run it

Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.

Provider	Speed	First token	Input $/M	Output $/M	Context	Quantisation
Azure	79 tok/s	1.67 s	$1.91	$3.83	1.0M	–
Relace	76 tok/s	0.89 s	$0.900	$1.50	1.0M	fp4
DeepInfra	68 tok/s	0.88 s	$1.30	$2.60	1.0M	fp8
NovitaAI	66 tok/s	2.15 s	$1.60	$3.20	1.0M	fp8
Venice	60 tok/s	1.29 s	$1.65	$3.30	1M	–
Alibaba Cloud Int.	53 tok/s	1.88 s	$1.42	$2.83	1M	fp8
Alibaba Cloud Int.	49 tok/s	1.75 s	$1.56	$3.12	1M	fp8
StreamLake	43 tok/s	1.33 s	$0.955	$1.91	1.0M	fp8
Baidu Qianfan	43 tok/s	1.29 s	$1.69	$3.38	1.0M	fp8
DigitalOcean	42 tok/s	1.17 s	$1.04	$2.09	1.0M	–
Cloudflare	42 tok/s	1.64 s	$1.15	$2.55	1.0M	–
SiliconFlow	41 tok/s	1.31 s	$1.50	$3.13	1.0M	fp8
AtlasCloud	40 tok/s	1.43 s	$1.68	$3.38	1.0M	fp4
Reka AI	38 tok/s	0.73 s	$1.05	$2.10	1.0M	–
GMICloud	24 tok/s	2.59 s	$0.957	$1.91	1.0M	fp8
Parasail	24 tok/s	1.49 s	$0.450	$3.48	1.0M	fp8

Price history

Listed price per 1M tokens over time, as recorded by OpenRouter for the provider with the longest history.

input outputnow $0.245 in · $0.490 out

What a task costs

Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.022 per 1M. Reasoning tokens are not modelled.

Workload	Tokens in / out	Cost	With caching	Time
Chat reply	400 / 300	$0.0017	$0.0013	50.3 s
Summarize a 30-page report	12,000 / 600	$0.018	$0.0065	53.2 s
Code edit	6,000 / 1,500	$0.014	$0.0080	1.0 min
Agentic coding session	60,000 / 4,000	$0.095	$0.037	1.4 min
Structured extraction	2,000 / 200	$0.0034	$0.0015	49.3 s