DDeepSeek · Released Dec 1, 2025
DeepSeek V3.2
DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism...
- Top 27%Fiction.LiveBench
- Top 28%Aider Polyglot
- Released
- Dec 1 2025
- Context window
- 164K 66K out
- Input / 1M tokens
- $0.28 from $0.21
- Output / 1M tokens
- $0.42 from $0.31
Where to run it
12 providers compared
Cheapest: GMICloud at $0.21 / $0.31 per 1M input / output tokens — 93% less than the priciest option.
| Provider | Input | Output | Cache read | Context | Max output | Uptime | vs. best |
|---|---|---|---|---|---|---|---|
| GMICloudFP8 | $0.21 | $0.31 | $0.022 | 164K | 147K | 100%24h | Cheapest |
| DeepInfraFP4 | $0.26 | $0.38 | $0.13 | 164K | 16K | 100%24h | +24% |
| AtlasCloudFP8 | $0.26 | $0.38 | $0.13 | 164K | 147K | 100%24h | +24% |
| Venice | $0.27 | $0.39 | $0.13 | 160K | 33K | 100%24h | +27% |
| BaiduFP8 · no tool calling | $0.28 | $0.42 | $0.028 | 131K | 66K | 73.6%24h | +34% |
| SiliconFlowFP8 | $0.26 | $0.80 | $0.14 | 164K | 147K | 100%24h | +68% |
| DigitalOceanno tool calling | $0.30 | $0.96 | $0.09 | 164K | 147K | 99.7%24h | +99% |
| Friendli | $0.50 | $1.50 | $0.25 | 164K | 147K | 97.2%24h | +220% |
| $0.56 | $1.68 | — | 164K | 66K | 99.9%24h | +259% | |
| Phala | $1.00 | $1.00 | $0.50 | 164K | 64K | 100%24h | +327% |
| SambaNovano tool calling | $3.00 | $4.50 | — | 33K | 7K | 92.7%24h | +1341% |
| Marano tool calling | $3.00 | $4.50 | — | 33K | 7K | 79.2%24h | +1341% |
Value
Price vs. performance
DeepSeek V3.2 Other DeepSeek Other labs Value frontier · Terminal-Bench vs. input price per 1M tokens (log scale) · Drag to pan, pinch or Ctrl + scroll to zoom, double-click to zoom in.
Benchmarks
11 benchmark scores
| Benchmark | Score | Rank | Leader |
|---|---|---|---|
| Terminal-BenchSoftware engineering · Tasks completed with the programs in a real terminal (best agent) | 39.6% | #23of 40 | GPT-5.584.7% |
| SciCodeSoftware engineering · Research-coding problems from the natural sciences | 38.9% | #77of 93 | Claude Opus 5.566.9% |
| Aider PolyglotSoftware engineering · Hard Exercism problems in many languages | 74.2% | #3of 11 | GPT-588.0% |
| APEX-AgentsAgents & computer use · Long professional tasks in banking, consulting and law | 21.3% | #49of 51 | Gemini 4 Argon82.4% |
| Vending-Bench 2Agents & computer use · Run a simulated vending business for a year (final balance) | $1,034 | #47of 61 | GPT-6 Astra$15,515 |
| Fiction.LiveBenchWriting & content · Understanding long creative writing (16k tokens) | 83.3% | #4of 15 | GPT-597.2% |
| WebDev ArenaDesign & visual · People vote on which model builds the better website (Elo) | 1367 | #72of 88 | Claude Opus 5.51820 |
| ARC-AGI-2Reasoning & knowledge · Few-shot abstract reasoning on grid puzzles | 4.0% | #59of 63 | GPT-6 Astra95.0% |
| ARC-AGI-1Reasoning & knowledge · Pattern induction from input–output examples | 57.0% | #53of 63 | Gemini 4 Argon98.5% |
| CritPtReasoning & knowledge · Open-ended, research-level physics problems | 2.9% | #65of 96 | GPT-5.6 Sol32.3% |
| LMCAReasoning & knowledge · Judge the quality of conceptual arguments like experts | 29.1% | #81of 96 | Claude Sonnet 5.568.3% |
Specs
Details
- Model ID
deepseek/deepseek-v3.2- Developer
- DeepSeek
- Release date
- Dec 1, 2025 ·
- Context window
- 163,840 tokens
- Max output
- 65,536 tokens
- API price
- $0.28 / $0.42 per 1M input / output tokens
- Providers
- 12
- Input types
- text
- Reasoning
- Yes
- Tool calling
- Yes
- Open weights
- Yes
- Free tier
- No
Release date, prices and providers from OpenRouter's catalog; buying directly from a provider can cost a little differently. Uptime is the share of OpenRouter requests each provider served over the last day. Benchmark scores are the best result per benchmark from Epoch AI, licensed CC BY 4.0. Rank counts every model tracked here with a score.