ModelPriceWatch.com
Last scan 2026-10-19 Models tracked 293 Providers 40 Cheapest paid Granite 4.0 H Micro $0.017/Mtok in Every price links to its source

DeepSeek V4 Flash

by DeepInfra

Current budget cheap tier

Today's price · per 1M tokens

Input

$0.060

per 1M tokens

Output

$0.180

per 1M tokens

Blended

$0.090

blended $/1M — 3:1 weighted input:output

Cached input

$0.015

25% of input — prompt caching

How this price is scoped: DeepInfra lists two entries for this model: DeepSeek-V4-Flash-0731 at $0.06 input / $0.015 cached input / $0.18 output per 1M tokens, and the undated DeepSeek-V4-Flash at $0.09 / $0.018 / $0.18. The tracked figure is the 0731 entry: DeepInfra describes it as the official release of DeepSeek-V4-Flash that supersedes the preview, DeepSeek's own API moved deepseek-v4-flash to the 0731 version on 2026-07-31, and it is the cheaper of the two. Call it as deepseek-ai/DeepSeek-V4-Flash-0731; the undated name costs 50% more on input.

Source: official DeepInfra pricing · read MODELPRICEWATCH.COM · 2026-10-19

Price receipt

We read DeepInfra's own pricing page on and found DeepSeek-V4-Flash-0731 listed with a price on the same row — that read is where the number above comes from, recorded as a price drop. We keep the page text we read; its content hash is d43a5326d8.

Overview

DeepSeek V4 Flash served on DeepInfra at $0.06/$0.18 per 1M tokens — the lowest per-token rate among tracked hosts (vs Fireworks $0.14/$0.28). 1M-token context.

Capabilities

struck through = not supported
Input 1/4
Text ✓ Image Audio Video
Output
Text ✓
Features 3/9
Prompt caching ✓ Reasoning Coding Fast inference ✓ Long context ✓ Open weights Multimodal Web search Realtime

Benchmark performance

accuracy % · higher is better
Avg benchmark score77.5%
Perf per $/Mtok861.1
GPQA Diamond
88.1%
SWE-Bench Verified vendor
79%
LiveCodeBench
91.6%
MCP Atlas
69%
BrowseComp
85.9%
Humanity's Last Exam
51.6%
ARC-AGI 2 vendor
18%
AIME 2025 vendor
58%
MMMLU vendor
75%
BFCL vendor
62%
HumanEval
79%
MATH 500 vendor
68%
Independent composite scores
  • AA Intelligence Index: 34.3 (Artificial Analysis composite across reasoning, knowledge and coding evals)
  • AA-Omniscience Index: -14.3 (−100–100)
  • GDPval-AA v2: 1433.7 ELO (Real-world work tasks, human baseline = 1000)
How it stacks up
  • Ranks #20 of 101 benchmarked models by average score
  • Ranks #63 of 195 comparably-measured models by percentile score, across 9 independent measurements
  • Ranks #2 of 195 comparably-tested models by normalized performance per dollar top value
  • Strongest at LiveCodeBench — 91.6%, #7 of 37

107 tokens/sec output

Source: Artificial Analysis, Vellum LLM Leaderboard, DeepSeek model card · updated · See full rankings →

Specifications

Provider
DeepInfra
Context window
1M tokens
Modality
text
Parameters
Proprietary
Open source
No — proprietary
Released
Status
Current
Last updated
Tags
budgetcheapfast

Availability verified: — listed on DeepInfra's own page