Pricing, context window, capabilities, and release date — pulled from each provider's public docs. Both are available via the same AIgateway OpenAI-compatible endpoint; flip the model string to switch.
Both models stream in parallel through your own AIgateway key. Tokens, latency, and cost update as they arrive.
| Deepseek-Math-7b-Instruct deepseek/deepseek-math-7b-instruct | Deepseek-Coder-6.7b-Base-Awq hf/thebloke/deepseek-coder-6.7b-base-awq | |
|---|---|---|
| Provider | DeepSeek | Hugging Face |
| Family | DeepSeek | DeepSeek |
| Modality | text | text |
| Context window | 4,096 tok | 4,096 tok |
| Max output | 4,096 tok | 4,096 tok |
| Released | 2024-02-27 | 2023-11-03 |
| Input price | — | $0.050 /1M |
| Output price | — | $0.100 /1M |
| Cache read | — | — |
| Tools | — | — |
| Streaming | yes | yes |
| Vision | — | — |
| JSON mode | — | — |
| Reasoning | — | — |
| Prompt caching | — | — |
DeepSeekMath-Instruct 7B is a mathematically instructed tuning model derived from DeepSeekMath-Base 7B. DeepSeekMath is initialized with DeepSeek-Coder-v1.5 7B and continues pre-training on math-related tokens sourced from Common Crawl, together with natural language and code data for 500B tokens.
Deepseek Coder is composed of a series of code language models, each trained from scratch on 2T tokens, with a composition of 87% code and 13% natural language in both English and Chinese.
# pip install aigateway-py openai
# aigateway-py: sub-accounts, evals, replays, jobs, webhook verify.
# openai SDK: chat/embeddings/images/audio — drop-in compat per our SDK's own guidance.
from openai import OpenAI
client = OpenAI(
base_url="https://api.aigateway.sh/v1",
api_key="sk-aig-...",
)
# Try Deepseek-Math-7b-Instruct
client.chat.completions.create(
model="deepseek/deepseek-math-7b-instruct",
messages=[{"role":"user","content":"hello"}],
)
# Try Deepseek-Coder-6.7b-Base-Awq — same client, same key
client.chat.completions.create(
model="hf/thebloke/deepseek-coder-6.7b-base-awq",
messages=[{"role":"user","content":"hello"}],
)