DeepSeek V3.2
The 35th most-used model on the router over the last 7 days, with 0.37% of tokens. Weekly share peaked at 6.8% in the week of Feb 2, 2026, with 28T tokens processed since Dec 1, 2025.
- Launch
- Dec 1, 2025
- Context
- 160k tokens
- Modality
- text → text
- Reasoning
- yes
- License
- Open weights
- Input
- $0.269 / 1M
- Output
- $0.40 / 1M
Weekly token share
Share of the router's weekly volume, since the first week with volume
See the numbers
| Week | Share |
|---|---|
| Aug 31, 2026 | 0.393% |
| Aug 24, 2026 | 0.385% |
| Aug 17, 2026 | 0.454% |
| Aug 10, 2026 | 0.546% |
| Aug 3, 2026 | 0.573% |
| Jul 27, 2026 | 0.707% |
| Jul 20, 2026 | 0.647% |
| Jul 13, 2026 | 0.561% |
| Jul 6, 2026 | 0.725% |
| Jun 29, 2026 | 0.955% |
| Jun 22, 2026 | 1.571% |
| Jun 15, 2026 | 2.119% |
| Jun 8, 2026 | 2.572% |
| Jun 1, 2026 | 3.232% |
| May 25, 2026 | 3.272% |
| May 18, 2026 | 3.387% |
| May 11, 2026 | 3.811% |
| May 4, 2026 | 3.370% |
| Apr 27, 2026 | 3.807% |
| Apr 20, 2026 | 5.826% |
| Apr 13, 2026 | 6.251% |
| Apr 6, 2026 | 6.036% |
| Mar 30, 2026 | 4.399% |
| Mar 23, 2026 | 5.467% |
| Mar 16, 2026 | 5.586% |
| Mar 9, 2026 | 6.140% |
| Mar 2, 2026 | 5.622% |
| Feb 23, 2026 | 5.875% |
| Feb 16, 2026 | 5.340% |
| Feb 9, 2026 | 5.811% |
| Feb 2, 2026 | 6.779% |
| Jan 26, 2026 | 6.150% |
| Jan 19, 2026 | 5.279% |
| Jan 12, 2026 | 4.644% |
| Jan 5, 2026 | 4.849% |
| Dec 29, 2025 | 6.327% |
| Dec 22, 2025 | 5.779% |
| Dec 15, 2025 | 3.518% |
| Dec 8, 2025 | 2.202% |
| Dec 1, 2025 | 1.166% |
Debuted in the week of Dec 1, 2025 and peaked at 6.8% 9 weeks later. The latest full week came in at 0.39%, 6% of peak.
What this model is used for new
Tasks where the model ranks among the leaders, and its share of each
Ranks among the leaders in 2 of the 29 classified tasks. Its largest share is in roleplay and fiction, 5.6% of the task's tokens.
Snapshot of the rolling 7-day window through Sep 10, 2026. Share: the model's tokens in the task over the task's total tokens. Weight: the task's share of classified volume.
Where to run it
Endpoints in the source's catalog, from cheapest to most expensive on input
| Provider | Input | Output | Context | Quantization | Zero retention |
|---|---|---|---|---|---|
| GMICloud | $0.209 | $0.31 | 160k | fp8 | no |
| StreamLake | $0.214 | $0.322 | 125k | fp8 | no |
| DigitalOcean | $0.25 | $0.80 | 160k | not reported | yes |
| SiliconFlow | $0.259 | $0.42 | 160k | fp8 | yes |
| DeepInfra | $0.26 | $0.38 | 160k | fp4 | yes |
| AtlasCloud | $0.26 | $0.38 | 160k | fp8 | no |
| Novita | $0.269 | $0.40 | 160k | fp8 | yes |
| Baidu | $0.28 | $0.42 | 128k | fp8 | no |
| Venice | $0.33 | $0.48 | 156k | not reported | yes |
| Alibaba | $0.37 | $1.11 | 128k | fp8 | no |
| Friendli | $0.50 | $1.50 | 160k | not reported | no |
| $0.56 | $1.68 | 160k | not reported | yes | |
| Phala | $1.00 | $1.00 | 160k | not reported | yes |
| Mara | $3.00 | $4.50 | 32k | not reported | yes |
| SambaNova | $3.00 | $4.50 | 32k | not reported | yes |
15 endpoints across 15 providers, with input from $0.209 to $3.00 per 1M tokens. 9 offer zero data retention.
Prices in US dollars per million tokens, from the latest archived endpoint catalog.
Benchmarks
Artificial Analysis indexes and benchmarks run by the source, with cost per task
GPQA Diamond 82.0%
$0.0041 per task, 5346 tasks. 58th of 131 evaluated. Median of evaluated models: 80.2%, $0.024 per task.
τ-bench verified, airline 72.1%
$0.023 per task, 1446 tasks. 50th of 122 evaluated. Median of evaluated models: 70.7%, $0.056 per task.
Artificial Analysis indexes, exposed by the catalog; benchmarks run by the source and archived on Sep 11, 2026. The gray mark is the median of evaluated models.
Against the week's largest models
The five largest proprietary and five largest open-weights models over the last 7 days
| Model | Lab | 7-day share | Blended price | Context | Intelligence | Weeks in top 10 |
|---|---|---|---|---|---|---|
| Proprietary | ||||||
| GPT-5.6 LunaOpenAI | OpenAI | 11.5% | $0.45 | 1M | 37.5 | 6 |
| Gemini 3.8 FlashGoogle | 2.1% | $1.50 | 1M | 41.2 | 0 | |
| Muse Spark 1.3 ContributorMeta | Meta | 1.5% | $0.125 | 1M | — | 0 |
| Claude Opus 5Anthropic | Anthropic | 1.4% | $10.00 | 1M | 50.7 | 1 |
| GPT-5.6 SolOpenAI | OpenAI | 1.4% | $4.00 | 1M | 47.1 | 0 |
| Open weights | ||||||
| Hy4 previewTencent | Tencent | 15.5% | $1.25 | 1M | — | 1 |
| DeepSeek V4 Flash 0731DeepSeek | DeepSeek | 10.0% | $0.094 | 1.31M | 34.5 | 5 |
| GLM 5.3 FlashZ.ai (GLM) | Z.ai (GLM) | 10.0% | $0.237 | 1.31M | 41.9 | 2 |
| MiMo-V2.5Xiaomi | Xiaomi | 4.2% | $0.175 | 1M | 22.3 | 15 |
| DeepSeek V4 Flash 0423DeepSeek | DeepSeek | 3.9% | $0.111 | 1M | 24.8 | 19 |
| DeepSeek V3.2this pageDeepSeek | DeepSeek | 0.37% | $0.302 | 160k | — | 26 |
Share over the last 7 days (Sep 4, 2026 to Sep 10, 2026). Blended price in US dollars per million tokens, weighted 75% input and 25% output. A dash means the number is not published, not zero.