OpenBMB
MiniCPM5-2B
Dense 2B-class open-source model for on-device and resource-constrained use, with native long-context support, tool calling, and agentic tasks
Model facts
- Context window
- 131K tokens
- Maximum output
- 131K tokens
- Cheapest paid input
- $0.124 per 1M tokens
- Cheapest paid output
- $0.743 per 1M tokens
- Open weights
- yes
- Providers
- 1
Capabilities and modalities
reasoning, tool calling, temperature control, open weights, text.
Local hardware estimate
At 4K context: Q4_K_M 2.5 GiB, Q8_0 3.7 GiB, F16 5.9 GiB working memory.
- Parameters
- 2.5 billion
- Architecture
- llama
- Model license
- apache-2.0
Planning estimate adapted from the MIT-licensed whichllm estimator using metadata from Hugging Face; not a vendor minimum requirement.
Artificial Analysis benchmarks
Compare this model on the full LLM benchmark leaderboard.
| Configuration | AAI | Coding | Math | Output tokens/s |
|---|---|---|---|---|
| default | 13.1 | 14.5 | not measured | not measured |
Observed price history
- 2026-09-14: $0.124 input / $0.743 output per 1M tokens
API providers
- AMD — model id MiniCPM5-2B; $0.124 input / $0.743 output per 1M tokens; provider documentation