DeepSeek V3.2 Speciale pricing
Source: litellm · updated 2026-11-06
Short answer: DeepSeek V3.2 Speciale costs $0.280 per million input tokens and $0.400 per million output tokens. That is a 3:1 blended rate of $0.310 per million tokens, ranking it #336 of 814 tracked models by input price.
Input / 1M
$0.280
Output / 1M
$0.400
Output ÷ input
1.43×
Context window
163,840
Price breakdown
| Metric | USD |
|---|---|
| Input, per 1M tokens | $0.280 |
| Output, per 1M tokens | $0.400 |
| Output ÷ input multiple | 1.43× |
| 3:1 blended, per 1M tokens | $0.310 |
| Context window | 163,840 tokens |
| Vendor | gmi |
What a typical workload costs
| Monthly volume | Input cost | Output cost | Total |
|---|---|---|---|
| 1M in / 250K out | $0.280 | $0.100 | $0.380 |
| 10M in / 2.5M out | $2.80 | $1.00 | $3.80 |
| 100M in / 25M out | $28.00 | $10.00 | $38.00 |
Other gmi models
| Model | Input | Output |
|---|---|---|
| Qwen 3 235 B A22 B Instruct 2507 Fp 8 | $0.087 | $0.350 |
| DeepSeek R1 Distill Qwen 7 B | $0.100 | $0.200 |
| Nvidia Nemotron 3.5 Super VL 120 B A12 B Bf 16 | $0.100 | $0.500 |
| GPT-4 O Mini | $0.150 | $0.600 |
| Nvidia Nemotron 3.5 Lightning 30 B A3 B Bf 16 | $0.150 | $1.00 |
Compare against
- DeepSeek V3.2 Speciale vs Qwen 3 VL 30 B A3 B Thinking
- DeepSeek V3.2 Speciale vs Qwen 3.7 Plus Variant 1
Cheaper alternatives
Next cheaper option: GLM 4.5 Air at $0.130 per 1M input tokens. Next step up: Us Gov Openai GPT OSS 120 B1.0 at $0.180 per 1M input tokens.
Spotted a price that is out of date? Submit a correction.