DeepSeek: R1 Distill Llama 70B vs Google: Gemma 4 31B
Side-by-side specification comparison of DeepSeek: R1 Distill Llama 70B and Google: Gemma 4 31B, including listed pricing, context window, capability flags, and output limits.
Catalog source
A dated routing-specification snapshot
AIViewer refreshed these two records on July 11, 2026 from OpenRouter's public model listings. The table reports routed prices and metadata, not first-party provider pricing or an independent performance test. Confirm availability and cost with the service you intend to use.
Side-by-side
Technical specifications
| DeepSeek: R1 Distill Llama 70B | Google: Gemma 4 31B | |
|---|---|---|
| Provider | DeepSeek | |
| Input price | $0.80/M | $0.12/M |
| Output price | $0.80/M | $0.35/M |
| Context window | 128K | 262K |
| Max output | 8K | 262K |
| Capabilities | Reasoning | Tool useVisionReasoning |
| Released | Jan 23, 2025 | Apr 2, 2026 |
Method and limits
What this comparison measures
This page compares fields in AIViewer's dated model-catalog snapshot: listed token prices, context and output limits, release dates, and capability flags. The readout is automated from those fields. It does not measure answer quality, latency, reliability, safety, or performance on your own workload. Prices and availability can change, so confirm important decisions with the provider before purchasing or deploying a model.
Read our evaluation and evidence standards