DeepSeek V3.2 vs DeepSeek V4 Pro 0813
At a Glance
| Compare |  DeepSeek V3.2DeepSeek | DeepSeek V4 Pro 0813DeepSeek |
|---|---|---|
| Intelligence, Cost, and Efficiency | ||
| IntelligenceHigher is better · MM Intelligence v2.5 | UnrankedNot in the 46-model eligible cohort | #20 of 4665.0 score · 3/3 sources · complete |
| CostLower is better · Published-token output estimate | UnrankedNot in the 44-model eligible cohort | #20 of 44$0.094 per LiveBench case |
| EfficiencyHigher is better · MM Efficiency v1.5 | UnrankedNot in the 38-model eligible cohort | #14 of 3857.4 score · 3/3 sources · complete |
| Pricing and Limits | ||
| Input priceFrom · USD / 1M tokens | $0.26Deepinfra ↗ · Sep 21, 2026 | $1.30Deepinfra ↗ · Sep 21, 2026 |
| Output priceFrom · USD / 1M tokens | $0.38Deepinfra ↗ · Sep 21, 2026 | $2.60Deepinfra ↗ · Sep 21, 2026 |
| Context windowMaximum documented tokens | 164K | 1,049K |
| Model facts checked | Aug 28, 2026View model evidence → | Sep 3, 2026View model evidence → |
Token prices are the lowest available sourced USD rates; input and output may use different providers. Cost ranking estimates output spend on LiveBench, not a full request bill. Ranking methodology →
Available Benchmarks
| Benchmark | DeepSeek-V3.2 | DeepSeek-V4-Pro-0813 |
|---|---|---|
| LMArena Text Arenatext-2026-09-13-d25aabda0010 · arena_rating · leader | 1,424.8099% of row best · rating · deepseek-v3.2; 95% CI [1421.27271719, 1428.32733406]; votes 46458; rank 96 | 1,443.95100% of row best · rating · deepseek-v4-pro-high-20260813; 95% CI [1437.16050212, 1450.73220317]; votes 9008; rank 58 |
| Overall ResultCounted from the protocol-matched rows above | 0 benchmark winsNo overall winner | 1 benchmark winNo overall winner |
Third-party benchmark Only like-for-like primary-publisher results are shown; raw scores, relative scores, configuration, and token spend remain visible.
Side-by-Side Facts
| Field | DeepSeek-V3.2 | DeepSeek-V4-Pro-0813 |
|---|---|---|
| Developer | DeepSeek | DeepSeek |
| Family | Deepseek V3 2 | Deepseek V4 Pro |
| Model | DeepSeek-V3.2 | DeepSeek-V4-Pro-0813 |
| Version | DeepSeek-V3.2 | DeepSeek-V4-Pro-0813 |
| Lifecycle | active | active |
| Released | 2025-12-01 | 2026-08-13 |
| Knowledge cutoff | Unknown | Unknown |
| Input modalities | Text | Text |
| Output modalities | Text | Text |
| Context window | 164K | 1,049K |
| Total parameters | 685.4B | 1.6T |
| Active parameters | Unknown | 49B |
| License | mit | mit |
| Open weights | Yes | Yes |
| API available | Yes | Yes |
| Self-hostable | Yes | Yes |
| Provider access | Deepinfra (Standard), Hugging Face (Standard), Openrouter (Standard) | Deepinfra (Standard), Fireworks Ai (Standard), Together Ai (Standard) |
| Capabilities | chat, generation, reasoning | agents, chat, coding, generation, reasoning, tools |
DeepSeek V3.2 Capabilities
DeepSeek V4 Pro 0813 Capabilities
Related Comparisons
All text comparisons →| A | Pair | B | Context |
|---|---|---|---|
DeepSeek-V3.2DeepSeek | vs |  GPT-6 AstraOpenAI | cross-developer peerstext |
Claude Fable 5.1Anthropic | vs | DeepSeek-V3.2DeepSeek | cross-developer peerstext |
DeepSeek-V3.2DeepSeek | vs |  Gemini 3.1 ProGoogle DeepMind | cross-developer peerstext |
DeepSeek-V3.2DeepSeek | vs | DeepSeek-V4.1-FlashDeepSeek | family variantstext |
DeepSeek-V3.2DeepSeek | vs | Grok 4.6xAI | cross-developer peerstext |
DeepSeek-V3.2DeepSeek | vs |  Qwen3.8-MaxQwen | cross-developer peerstext |
DeepSeek-V3.2DeepSeek | vs |  Kimi-K3Moonshot AI | cross-developer peerstext |
 MiniMax-M3MiniMax | vs | DeepSeek-V3.2DeepSeek | cross-developer peerstext |
DeepSeek-V3.2DeepSeek | vs |  GLM-5.3Z.ai | cross-developer peerstext |
DeepSeek-V3.2DeepSeek | vs |  Hy4 previewTencent | cross-developer peerstext |
Seed 2.1 ProByteDance Seed | vs | DeepSeek-V3.2DeepSeek | cross-developer peerstext |
DeepSeek-V3.2DeepSeek | vs |  Mistral Large 3Mistral AI | cross-developer peerstext |
Primary Evidence
Sources and Freshness
Questions
DeepSeek V3.2 vs DeepSeek V4 Pro 0813 FAQs
Is DeepSeek V3.2 or DeepSeek V4 Pro 0813 better for coding?+
This comparison does not currently contain a protocol-matched coding benchmark for both DeepSeek V3.2 and DeepSeek V4 Pro 0813, so Model Markets cannot name a coding leader from pricing, context size, or capability labels alone.
Which is cheaper, DeepSeek V3.2 or DeepSeek V4 Pro 0813?+
DeepSeek V3.2 is $0.26 and DeepSeek V4 Pro 0813 is $1.30 per million tokens, so DeepSeek V3.2 is cheaper on this metric. DeepSeek V3.2 is $0.38 and DeepSeek V4 Pro 0813 is $2.60 per million tokens, so DeepSeek V3.2 is cheaper on this metric.
Which has a larger context window, DeepSeek V3.2 or DeepSeek V4 Pro 0813?+
DeepSeek V4 Pro 0813 has the larger sourced context window. DeepSeek V3.2 supports 164K and DeepSeek V4 Pro 0813 supports 1,049K.
Which performs better in benchmarks, DeepSeek V3.2 or DeepSeek V4 Pro 0813?+
There is no overall benchmark winner: At least two independently verified, protocol-matched benchmarks are required for an overall winner.
Can DeepSeek V3.2 or DeepSeek V4 Pro 0813 be self-hosted?+
Both models have the same recorded self-hosting status: supported. DeepSeek V3.2 is open weight; DeepSeek V4 Pro 0813 is open weight.
Can DeepSeek V3.2 and DeepSeek V4 Pro 0813 understand images?+
DeepSeek V3.2 is not documented with image input; DeepSeek V4 Pro 0813 is not documented with image input. This reflects supported input modalities, not vision quality.
Which can generate longer answers, DeepSeek V3.2 or DeepSeek V4 Pro 0813?+
Neither has a larger sourced maximum output. DeepSeek V3.2 is — and DeepSeek V4 Pro 0813 is —.
Do DeepSeek V3.2 and DeepSeek V4 Pro 0813 support reasoning and tool use?+
DeepSeek V3.2: reasoning. DeepSeek V4 Pro 0813: reasoning and tool calling. Feature support does not establish relative quality.
Which is available from more inference providers, DeepSeek V3.2 or DeepSeek V4 Pro 0813?+
DeepSeek V3.2 has 3 sourced provider routes; DeepSeek V4 Pro 0813 has 3, a tie.
Which offers better value, DeepSeek V3.2 or DeepSeek V4 Pro 0813?+
There is no universal value winner. Compare the input and output prices above with the matched benchmark result for your workload: cheaper tokens can be offset by different quality, token usage, latency, or provider availability.