What are DeepSeek V4.1 Flash 1017's benchmark scores?

As of 25 Oct 2026, DeepSeek V4.1 Flash 1017 (max effort) scores 64.0 ±3.9 on the BenchLeader Index, top 100, strongest in composite (80) and weakest in knowledge (57). That rests on 16 independent benchmark results, and every score on this page links to its publisher.

How does DeepSeek V4.1 Flash 1017 rank?

DeepSeek V4.1 Flash 1017 (max effort) is #58 of 785 ranked configurations on 25 Oct 2026 (top 100). 133 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.

How much does DeepSeek V4.1 Flash 1017 cost per 1M tokens?

As of 25 Oct 2026, DeepSeek V4.1 Flash 1017 costs $0.300 per 1M input tokens and $1.20 per 1M output tokens, $0.525 blended (three input tokens per output token).

Is DeepSeek V4.1 Flash 1017 better than Muse Spark 1.3?

As of 25 Oct 2026, DeepSeek V4.1 Flash 1017 (max) and Muse Spark 1.3 (max) are neck and neck on quality (64.0 vs 67.1 on the BenchLeader Index, inside the ±5.9 margin of error), and DeepSeek V4.1 Flash 1017 (max) costs 3.8× less, which makes it the better buy for most workloads. The head-to-head page has the benchmark-by-benchmark detail.

When was DeepSeek V4.1 Flash 1017 released?

DeepSeek released DeepSeek V4.1 Flash 1017 on 17 Oct 2026. BenchLeader has results for it at 5 reasoning-effort settings.

DeepSeek Open weights: deepseek-ai/DeepSeek-V4.1-Flash-1017Reasoning modelCompare with another model

DeepSeek V4.1 Flash 1017

Reasoning effort

DeepSeek V4.1 Flash 1017 is a DeepSeek open-weights reasoning model, released 17 Oct 2026. Its best configuration (max reasoning effort) ranks #58 of 785 on the BenchLeader Index at 64.0 ±3.9, in the upper half. The ± is the point: 133 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 composite (80) and lowest in knowledge (57). At $0.525 per million tokens blended it is mid-priced. Output speed of 195 tokens per second puts it in the fastest quarter, with a first answer in 11.4 s. It has been measured at 5 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 23 Oct 2026.

Blended price: $0.525/M; $0.300 in · $1.20 out
Output speed: 195 tok/s; measured by Artificial Analysis
First answer: 11 s; first token 0.86 s
Context: 1.0M
Full answer: 14 s; median, reasoning included
Cost per run: $0.262; one full Intelligence Index run
Released: 17 Oct 2026

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

Overall index64
Reasoning71
Coding70
Agents & tools64
Maths62
Knowledge57
Multimodal64
Long context66
Composite80

Versions

DeepSeek has shipped 2 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.

Model	Released	Index	Rank
DeepSeek V4.1 Flash 1017maxthis page	17 Oct 2026	64.0	#58
DeepSeek V4.1 Flashmax	10 Sept 2026	61.2	#98

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	–	–	80 tok/s	0.86 s	$0.0005	Reasoning 30
low	–	–	80 tok/s	0.86 s	$0.0005	Reasoning 65
high	–	–	80 tok/s	0.86 s	$0.0005	Reasoning 64
maxbest	64.0	#58	195 tok/s	11 s	$0.0005	Agents & tools 64 · Coding 70 · Composite 80 · Knowledge 57 · Long context 66 · Maths 62 · Multimodal 64 · Reasoning 71
not stated	–	–	80 tok/s	0.86 s	$0.0005	Composite 63

How its index has moved

19 Oct 2026 to 24 Oct 2026

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#28	Artificial Analysis

Reasoning

Benchmark	no reasoning	low	high	max	not stated	Source
GPQA Diamond	–	–	–	93.2%#20	–	Epoch AI Benchmarking Hub
LiveBench 2.0 Reasoningnot in indexeffort not stated	–	–	–	–	89.1%#20	LiveBench
Humanity's Last Exam (AA)not in index	–	–	–	43.3%#60	–	Artificial Analysis
ARC-AGI-1	5.7%#242	91.5%#65	94.5%#39	93.5%#49	–	ARC Prize
ARC-AGI-2	2.1%#201	73.6%#52	68.6%#63	78.1%#42	–	ARC Prize
CritPt	–	–	–	24.0%#50	–	Artificial Analysis
Chess Puzzlesnot in index	–	–	–	46.0%#26	–	Epoch AI Benchmarking Hub
Mystery Game Puzzlesnot in index	–	–	–	49.0%#14	–	Epoch AI Benchmarking Hub

Coding

Benchmark	max	not stated	Source
Arena WebDev	1665#13	–	Arena (LMArena)
LiveBench 2.0 Codingnot in indexeffort not stated	–	96.0%#4	LiveBench
SciCode (AA)not in index	55.9%#44	–	Artificial Analysis
Code Migrationnot in index	44.2%#24	–	Vals AI
Code Migration v1.1not in index	67.5%#7	–	Vals AI
Vibe Code Bench v1.1not in index	87.7%#11	–	Vals AI
Terminal-Bench 4.0 (AA)not in index	33.8%#38	–	Artificial Analysis
FrontierCode 1.1 (AA)not in index	43.7%#19	–	Artificial Analysis

Agents & tools

Benchmark	max	not stated	Source
LiveBench Long Agenticnot in indexeffort not stated	–	50.9%#12	LiveBench
LiveBench Browsernot in indexeffort not stated	–	76.0%#20	LiveBench
GDPval-AA v2.1not in index	62.3%#8	–	Artificial Analysis
Finance Agent v2not in index	57.0%#18	–	Vals AI
Finance Agent v2.1not in index	62.9%#12	–	Vals AI
Harvey's Legal Agent Benchmark v1.1not in index	14.2%#9	–	Vals AI
Harvey's Legal Agent Benchmark v1not in index	9.2%#22	–	Vals AI
Legal Research Benchnot in index	45.2%#18	–	Vals AI
Tax Agent Benchnot in index	44.4%#6	–	Vals AI
Terminal-Bench 4.0 (Vals)not in index	31.8%#19	–	Vals AI
AutomationBenchnot in index	66.6%#14	–	Artificial Analysis
GDP.pdf (AA)not in index	17.2%#83	–	Artificial Analysis
AA-Briefcase v1.1not in index	1570#12	–	Artificial Analysis
Terminal-Bench-Science 0.1 (AA)not in index	12.9%#28	–	Artificial Analysis
CWE-Bench (AA)	69.2%#1	–	Artificial Analysis
DeepSec-Bench (AA)	24.9%#9	–	Artificial Analysis
CyberGym end-to-end (AA)	76.3%#5	–	Artificial Analysis

Maths

Benchmark	max	Source
FrontierMath Tiers 1–3	71.2%#27	Epoch AI Benchmarking Hub
FrontierMath Tier 4	46.3%#27	Epoch AI Benchmarking Hub
OTIS Mock AIME	99.4%#15	Epoch AI Benchmarking Hub

Knowledge

Benchmark	max	not stated	Source
SimpleQA Verified	42.5%#43	–	Epoch AI Benchmarking Hub
LiveBench 2.0 Data Analysisnot in indexeffort not stated	–	79.5%#12	LiveBench
LiveBench Researchnot in indexeffort not stated	–	79.5%#18	LiveBench
AA-Omniscience	4.7#109	–	Artificial Analysis
Excel Modeling Benchmarknot in index	61.8%#33	–	Vals AI
Excel Modeling Benchmark v1.1not in index	69.2%#16	–	Vals AI
AA-Omniscience: accuracynot in index	42.9%#105	–	Artificial Analysis
AA-Omniscience: non-hallucinationnot in index	33.3%#201	–	Artificial Analysis

Multimodal

Benchmark	max	not stated	Source	Trend
LiveBench Designnot in indexeffort not stated	–	83.4%#4	LiveBench
MMMU-Pro	80.6%#45	–	Artificial Analysis

Long context

Benchmark	max	Source	Trend
AA-LCR	82.7%#36	Artificial Analysis
MLCRnot in index	26.7%#23	Artificial Analysis

Composite

Benchmark	max	not stated	Source
LiveBench 2.0effort not stated	–	79.2%#10	LiveBench
AA Intelligence Index v4.3.2	45.1#37	–	Artificial Analysis
Vals Indexnot in index	61.3#9	–	Vals AI
Vals Finance Indexnot in index	60.9%#7	–	Vals AI
Vals Router Indexnot in indexeffort not stated	–	61.3%#8	Vals AI
AA Cyber Indexnot in index	56.8#2	–	Artificial Analysis

Where it wins

Benchmarks where this configuration ranks in the top five of every configuration measured.

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
Together	261 tok/s	0.30 s	$0.300	$1.20	1.0M	–
Fireworks Spot	201 tok/s	0.84 s	$0.210	$0.840	1.0M	–
Baseten	194 tok/s	0.50 s	$0.300	$1.20	1.0M	–
DigitalOcean	193 tok/s	0.79 s	$0.225	$0.900	1.0M	–
Fireworks	184 tok/s	0.78 s	$0.300	$1.20	1.0M	–
Wafer	181 tok/s	0.41 s	$0.130	$0.500	1.0M	mxfp4
Morph	172 tok/s	0.27 s	$0.411	$0.600	1.0M	fp8
Prime Intellect	171 tok/s	0.82 s	$0.300	$1.20	1.0M	–
Crusoe	139 tok/s	0.65 s	$0.290	$1.20	1.0M	fp8
Parasail	137 tok/s	0.40 s	$0.300	$1.20	1.0M	fp8
Nebius Token Factory	133 tok/s	0.49 s	$0.300	$1.20	1.0M	fp8
Modal	123 tok/s	0.76 s	$0.250	$1.00	1.0M	–
Relace	122 tok/s	0.91 s	$0.070	$0.500	1.0M	–
Decart	120 tok/s	0.77 s	$0.137	$0.546	1.0M	fp4
Phala	117 tok/s	1.32 s	$0.300	$1.20	1.0M	–
Reka AI	116 tok/s	0.53 s	$0.080	$0.800	1.0M	–
Makora	115 tok/s	0.67 s	$0.150	$1.20	1.0M	fp8
AtlasCloud	102 tok/s	1.11 s	$0.195	$0.780	1.0M	fp8
Baidu Qianfan	99 tok/s	1.32 s	$0.280	$1.13	1.0M	fp8
Ionstream	98 tok/s	0.51 s	$0.220	$0.600	1.0M	–
inference.net	96 tok/s	0.40 s	$0.080	$0.500	1.0M	fp8
Alibaba Cloud Int.	95 tok/s	0.72 s	$0.330	$1.32	1M	–
Cloudflare	90 tok/s	1.49 s	$0.100	$0.600	1.0M	–
SiliconFlow	89 tok/s	1.41 s	$0.300	$1.20	1.0M	fp8
inference.net	87 tok/s	0.36 s	$0.080	$0.500	1.0M	fp8
DeepSeek	85 tok/s	0.81 s	$0.150	$0.600	1.0M	–
Open Inference	73 tok/s	0.89 s	$0.020	$1.00	1.0M	fp4
NovitaAI	73 tok/s	1.57 s	$0.240	$0.960	1.0M	fp8
DeepInfra	70 tok/s	1.34 s	$0.147	$0.441	1.0M	fp8
Sail Research	69 tok/s	1.09 s	$0.101	$0.150	1.0M	fp4
DeepInfra	68 tok/s	0.78 s	$0.098	$0.294	1.0M	fp8
Sail Research	66 tok/s	1.23 s	$0.101	$0.150	1.0M	fp4
GMICloud	64 tok/s	1.96 s	$0.180	$0.720	1.0M	fp8
DeepInfra	57 tok/s	3.03 s	$0.078	$0.235	1.0M	fp8

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 $0.006 per 1M. Reasoning tokens are not modelled.

Workload	Tokens in / out	Cost	With caching	Time
Chat reply	400 / 300	$0.0005	$0.0004	12.9 s
Summarize a 30-page report	12,000 / 600	$0.0043	$0.0017	14.5 s
Code edit	6,000 / 1,500	$0.0036	$0.0023	19.1 s
Agentic coding session	60,000 / 4,000	$0.023	$0.0096	31.9 s
Structured extraction	2,000 / 200	$0.0008	$0.0004	12.4 s

New this month

Compare DeepSeek V4.1 Flash 1017 with…

Questions people ask

What are DeepSeek V4.1 Flash 1017's benchmark scores?: As of 25 Oct 2026, DeepSeek V4.1 Flash 1017 (max effort) scores 64.0 ±3.9 on the BenchLeader Index, top 100, strongest in composite (80) and weakest in knowledge (57). That rests on 16 independent benchmark results, and every score on this page links to its publisher.
How does DeepSeek V4.1 Flash 1017 rank?: DeepSeek V4.1 Flash 1017 (max effort) is #58 of 785 ranked configurations on 25 Oct 2026 (top 100). 133 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
How much does DeepSeek V4.1 Flash 1017 cost per 1M tokens?: As of 25 Oct 2026, DeepSeek V4.1 Flash 1017 costs $0.300 per 1M input tokens and $1.20 per 1M output tokens, $0.525 blended (three input tokens per output token).
Is DeepSeek V4.1 Flash 1017 better than Muse Spark 1.3?: As of 25 Oct 2026, DeepSeek V4.1 Flash 1017 (max) and Muse Spark 1.3 (max) are neck and neck on quality (64.0 vs 67.1 on the BenchLeader Index, inside the ±5.9 margin of error), and DeepSeek V4.1 Flash 1017 (max) costs 3.8× less, which makes it the better buy for most workloads. The head-to-head page has the benchmark-by-benchmark detail.
When was DeepSeek V4.1 Flash 1017 released?: DeepSeek released DeepSeek V4.1 Flash 1017 on 17 Oct 2026. BenchLeader has results for it at 5 reasoning-effort settings.

Data as of 25 Oct 2026. Compare these configurations.

Cite as: BenchLeader, “DeepSeek V4.1 Flash 1017: benchmarks, pricing, speed and rank”, data as of 25 Oct 2026, https://www.benchleader.com/models/deepseek-v4-1-flash-1017.