Stable Code Instruct 3B
Stable Code Instruct 3B is a 3B parameter code-completion model from Stability AI, released January 2024 and trained on permissively licensed Stack data with 16K context for whole-file fill-in-middle tasks. At the time of release it was a reasonable lightweight option for single-file completions on constrained hardware. By mid-2026 it has been effectively displaced for serious deployments: Qwen 2.5 Coder 7B and DeepSeek Coder offer materially stronger benchmark scores at comparable or lower hosted cost. The license is Stability AI's Non-Commercial Community License — commercial use requires a Stability membership, which adds friction most teams won't accept when permissively licensed alternatives exist. Treat it as a research artifact or a local fine-tuning base, not a production endpoint.
No pricing data available for this model yet.
Not enough history to display chart. View full history →
No benchmark data available.
Questions developers ask
How much does it cost to run Stable Code Instruct 3B for 100M tokens?▾
Pricing varies by provider. Use the workload calculator to get an accurate estimate for your specific usage pattern.
What is the cheapest provider for Stable Code Instruct 3B?▾
No pricing data is available yet. Check back after the next scrape run.
What context window does Stable Code Instruct 3B support?▾
Stable Code Instruct 3B supports a context window of 16,384 tokens. Individual providers may cap this lower.
What's the cheapest way to run Stable Code Instruct 3B?▾
Compare provider prices in the table above and consider using prompt caching if supported.
Is there a free tier for Stable Code Instruct 3B?▾
Free tiers vary by provider and change frequently. Check each provider's current pricing page for trial credits or free-tier limits. The prices shown on this page reflect paid API access.
How much does Stable Code Instruct 3B cost for 1M tokens?▾
Pricing data for Stable Code Instruct 3B is not yet available. Check back after the next data refresh.
Keep exploring
No hosted pricing scraped for this model yet. Methodology · Raw data