DeepSeek V3.2
DeepSeek's V3.2 — open-weight frontier reasoning at a fraction of closed-source pricing.
ToolsStructured output
Context window
164K
Max output
20K
Input price
$0.56 /1M
Output price
$1.68 /1M
Benchmarks
Independent benchmark scores — composite indices for reasoning, coding, and math, plus individual eval scores where available.
Intelligence Index32.1
Coding Index34.6
Math Index59.0
MMLU-Pro
84%
GPQA Diamond
75%
Humanity's Last Exam
11%
LiveCodeBench
59%
SciCode
39%
AIME 2025
59%
IFBench
49%
Long-context reasoning
39%
Terminal-Bench Hard
33%
τ²-Bench Telecom
79%
See full leaderboard →Benchmarks via Artificial Analysis · View on AA
Available routes
The same DeepSeek V3.2 runs on 4 different routes through the Opper gateway. Each route has its own data-handling posture.
| Provider | Region | Zero data retention | Training | Logging | Contract | Input | Output |
|---|---|---|---|---|---|---|---|
| Fireworks | US | Default | No | None for open models by default — does not log or store prompt or generation data without explicit opt-in. | PAYG | $0.56 | $1.68 |
| Infercom | US | — | No | Pending verification | PAYG | $3.24 | $4.86 |
| Nebius | US | Opt-in | No | Per privacy policy: not detailed. | PAYG | $0.30 | $0.45 |
| Novita | US | None | — | Not published; appendix available upon request. | PAYG | $0.27 | $0.40 |
Human review: No human review