VerdictPal
Join

Models / DeepSeek

DeepSeek V4 Flash Max

DeepSeek V4 · Released 2026-04-24

DeepSeek's April 2026 cost-efficient V4 tier. 1M context, $0.14/$0.28 per 1M tokens (cache miss). SWE-bench Verified 79.0%.

Cheapest long-context coding route on the public ledger. Default DeepSeek API pick for agents.

Composite78
Context1M
Input / 1M$0.14
Output / 1M$0.28
Knowledge cutoffopen
Statusflagship
DeepSeek
Not enough families covered yet.Add more benchmark rows for DeepSeek V4 Flash Max to see the radar.
Benchmark placements

Where DeepSeek V4 Flash Max places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
BFCLTool use#13/ 2077Berkeley Function Calling Leaderboard V42026-04-15
SWE-bench VerifiedCoding#15/ 2379SWE-bench official leaderboard2026-05-30
Family context

The three highest-scoring carded models in each capability family. Where DeepSeek V4 Flash Max shows up, it's highlighted.

CodingDeepSeek V4 Flash Max · 79
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
Long context
  1. 1
    Kimi K3
    75
  2. 2
    GPT-5.1
    75
  3. 3
    GPT-5.6 Sol
    74
Tool useDeepSeek V4 Flash Max · 77
  1. 1
    Claude Sonnet 4.6
    89
  2. 2
    GPT 5.5
    87
  3. 3
    Claude Opus 4.8
    87
Newest receipts

Benchmark rows added to the public ledger for DeepSeek V4 Flash Max in the last 120 days. Older rows live in the full table below.

2 recent
BenchmarkFamilyScoreSourceDays ago
SWE-bench VerifiedCoding79%SWE-bench official leaderboard54
BFCLTool use76.8%Berkeley Function Calling Leaderboard V499
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Coding79
Tool use77
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover or tab any dot for name, score, and input price.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMiniMaxxAIMetaZhipuXiaomiMeituanCohere

Full benchmark ledger

Every catalog benchmark for DeepSeek V4 Flash Max. Scores link to the original source; gaps mean no public row exists yet.

2 sourced rows

2 of 32 catalog benchmarks have a sourced row for DeepSeek V4 Flash Max.

Coding

Tool use

Sources