Loading...
Loading...
Side-by-side comparison of API pricing, context window, latency, and benchmark performance. Data refreshed daily from provider APIs and ModelStop's own latency probes.
💰math-1b-sft-numinamath-bs512-from-step80000 is 100% cheaper than Qwen: Qwen2.5 Coder 7B Instruct
Price (lower is better), Speed/Latency (lower is better), Performance (higher is better)
math-1b-sft-numinamath-bs512-from-step80000 is Infinity% cheaper
| Spec | math-1b-sft-numinamath-bs512-from-step80000 | Qwen: Qwen2.5 Coder 7B Instruct |
|---|---|---|
| Provider | pre-to-post-olmo | qwen |
| Input price / 1M tokens | Free | $0.03 |
| Output price / 1M tokens | — | $0.09 |
| Context window | — | 33k tokens |
| Latency (p50) | — | — |
| Best benchmark score | — | — |
| Open source | No | No |
Open-source math-1b-sft-numinamath-bs512-from-step80000 model from pre-to-post-olmo — available for download and self-hosting on Hugging Face.
Qwen2.5-Coder-7B-Instruct is a 7B parameter instruction-tuned language model optimized for code-related tasks such as code generation, reasoning, and bug fixing. Based on the Qwen2.5 architecture, it incorporates enhancements like RoPE,...
Want to swap in a different model or estimate monthly costs?