ModelPriceWatch.com
Last scan 2026-11-13 Models tracked 315 Providers 42 Cheapest paid pplx-decider-flash-v1.3 $0.010/Mtok in Every price links to its source

DeepSeek V4 Flash

by Fireworks

Retired budget cheap tier
Retired on . No longer served on the endpoint we price — the figures below are kept for historical reference only.
Availability: Fireworks took DeepSeek V4 Flash (0731) off its serverless API on 2026-09-25. Its changelog announced the cut on 2026-09-12 and confirmed on 2026-09-26 that it is in effect, and the model's own page says the deprecation is "only on serverless": you can still run it on a dedicated on-demand deployment, which Fireworks bills by GPU time, not per token. Fireworks points serverless users to DeepSeek V4.1 Flash. The serverless pricing page no longer lists it, so $0.14/$0.28 per 1M tokens (cached $0.028) is what the model last cost per token on Fireworks, not a rate you can buy today.

Today's price · per 1M tokens

Input

$0.140

per 1M tokens

Output

$0.280

per 1M tokens

Blended

$0.175

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

Cached input

$0.028

20% of input — prompt caching

Source: official Fireworks pricing · read MODELPRICEWATCH.COM · 2026-11-13

Price receipt

We read Fireworks'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. We keep the page text we read; its content hash is d9947041e5.

DeepSeek V4 Flash is still sold by DeepInfra at $0.090 blended per 1M tokens — see the consolidated DeepSeek V4 Flash page →

Overview

DeepSeek V4 Flash on Fireworks, retired from serverless on 2026-09-25. Last listed at $0.14/$0.28 per 1M; cached $0.028.

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 score77.2%
Perf per $/Mtok441.1
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%
MATH 500 vendor
68%

Sources: Vellum LLM Leaderboard (GPQA Diamond, LiveCodeBench, MCP Atlas, BrowseComp, Humanity's Last Exam) · DeepSeek model card (SWE-Bench Verified, ARC-AGI 2, AIME 2025, MMMLU, BFCL, MATH 500)

Independent composite scores
  • GDPval-AA v2: 1436.8 ELO (Real-world work tasks, human baseline = 1000)

Source: Artificial Analysis (artificialanalysis.ai)

How it stacks up
  • Ranks #20 of 100 benchmarked models by average score
  • Ranks #53 of 189 comparably-measured models by percentile score, across 6 independent measurements
  • Ranks #3 of 189 comparably-tested models by normalized performance per dollar top value
  • Strongest at LiveCodeBench — 91.6%, #6 of 35

200 tokens/sec output

Throughput: Artificial Analysis via Vellum · Source: Artificial Analysis (artificialanalysis.ai)

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

Specifications

Provider
Fireworks
Context window
1M tokens
Modality
text
Parameters
Proprietary
Open source
No — proprietary
Released
Status
Retired ended
Last updated
Tags
budgetfastcachingretired

Availability verified: — per Fireworks's own deprecation notice