DeepSeek V3 vs DeepSeek V3.2
At a Glance
| Compare |  DeepSeek V3DeepSeek | DeepSeek V3.2DeepSeek |
|---|---|---|
| Pricing and Limits | ||
| Input priceFrom · USD / 1M tokens | $0.2574Openrouter ↗ · Aug 29, 2026 | $0.26Deepinfra ↗ · Oct 19, 2026 |
| Output priceFrom · USD / 1M tokens | $0.89Deepinfra ↗ · Oct 19, 2026 | $0.38Deepinfra ↗ · Oct 19, 2026 |
| Context windowMaximum documented tokens | 164K | 164K |
| Model facts checked | Aug 28, 2026View model evidence → | Aug 28, 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 | DeepSeek-V3.2 |
|---|---|---|
| LMArena Text Arenatext-2026-10-15-0ebb411abf00 · arena_rating · leader | 1,332.4694% of row best · rating · deepseek-v3; 95% CI [1327.72698428, 1337.19353550]; votes 21770; rank 224 | 1,424.46100% of row best · rating · deepseek-v3.2; 95% CI [1420.99052972, 1427.92299771]; votes 48112; rank 105 |
| 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 | DeepSeek-V3.2 |
|---|---|---|
| Developer | DeepSeek | DeepSeek |
| Family | Deepseek V3 | Deepseek V3 2 |
| Model | DeepSeek-V3 | DeepSeek-V3.2 |
| Version | DeepSeek-V3 | DeepSeek-V3.2 |
| Lifecycle | active | active |
| Released | 2024-12-26 | 2025-12-01 |
| Knowledge cutoff | Unknown | Unknown |
| Input modalities | Text | Text |
| Output modalities | Text | Text |
| Context window | 164K | 164K |
| Total parameters | 684.5B | 685.4B |
| Active parameters | 37B | Unknown |
| License | Unknown | mit |
| Open weights | Yes | Yes |
| API available | Yes | Yes |
| Self-hostable | Yes | Yes |
| Provider access | Deepinfra (Standard), Hugging Face (Standard), Openrouter (Standard) | Deepinfra (Standard), Hugging Face (Standard), Openrouter (Standard) |
| Capabilities | chat, generation | chat, generation, reasoning |
DeepSeek V3 Capabilities
DeepSeek V3.2 Capabilities
Related Comparisons
All text comparisons →| A | Pair | B | Context |
|---|---|---|---|
DeepSeek-V3DeepSeek | vs |  Sonar ProPerplexity | cross-developer peerstext |
DeepSeek-V3DeepSeek | vs | DeepSeek-V4-Flash-BaseDeepSeek | family variantstext |
DeepSeek-R1DeepSeek | vs | DeepSeek-V3DeepSeek | family variantstext |
DeepSeek-V3DeepSeek | vs |  Llama-3.1-405BMeta | cross-developer peerstext |
DeepSeek-V3DeepSeek | vs | cross-developer peerstext | |
DeepSeek-V3DeepSeek | vs | cross-developer peerstext | |
 MiniMax-M2.5MiniMax | vs | DeepSeek-V3DeepSeek | cross-developer peerstext |
DeepSeek-V3DeepSeek | vs |  Mistral Large 3Mistral AI | cross-developer peerstext |
DeepSeek-V3DeepSeek | vs |  Kimi-K2-InstructMoonshot AI | cross-developer peerstext |
DeepSeek-V3DeepSeek | vs |  GLM-5Z.ai | cross-developer peerstext |
DeepSeek-V3DeepSeek | vs | GLM-5.1Z.ai | cross-developer peerstext |
Primary Evidence
Sources and Freshness
Questions
DeepSeek V3 vs DeepSeek V3.2 FAQs
Is DeepSeek V3 or DeepSeek V3.2 better for coding?+
This comparison does not currently contain a protocol-matched coding benchmark for both DeepSeek V3 and DeepSeek V3.2, so Model Markets cannot name a coding leader from pricing, context size, or capability labels alone.
Which is cheaper, DeepSeek V3 or DeepSeek V3.2?+
DeepSeek V3 is $0.2574 and DeepSeek V3.2 is $0.26 per million tokens, so DeepSeek V3 is cheaper on this metric. DeepSeek V3 is $0.89 and DeepSeek V3.2 is $0.38 per million tokens, so DeepSeek V3.2 is cheaper on this metric.
Which has a larger context window, DeepSeek V3 or DeepSeek V3.2?+
Both models have the same sourced context window: 163,840 tokens each.
Which performs better in benchmarks, DeepSeek V3 or DeepSeek V3.2?+
There is no overall benchmark winner: At least two independently verified, protocol-matched benchmarks are required for an overall winner.
Can DeepSeek V3 or DeepSeek V3.2 be self-hosted?+
Both models have the same sourced self-hosting status: supported. DeepSeek V3 is open weight; DeepSeek V3.2 is open weight.
Can DeepSeek V3 and DeepSeek V3.2 understand images?+
DeepSeek V3 is not documented with image input; DeepSeek V3.2 is not documented with image input. This reflects supported input modalities, not vision quality.
Which can generate longer answers, DeepSeek V3 or DeepSeek V3.2?+
Neither model has a sourced maximum output in this comparison, so a larger limit cannot be determined.
Do DeepSeek V3 and DeepSeek V3.2 support reasoning and tool use?+
DeepSeek V3: none of these features are definitively sourced. DeepSeek V3.2: reasoning. Feature support does not establish relative quality.
Which is available from more inference providers, DeepSeek V3 or DeepSeek V3.2?+
DeepSeek V3 has 3 sourced provider routes; DeepSeek V3.2 has 3, a tie.
Which offers better value, DeepSeek V3 or DeepSeek V3.2?+
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.