Compare
R1 0528 vs Sonar
This page is context-first: how much text each model can take in one request. Full specs adds capabilities and limits; the pricing matrix below is only about $/million tokens from hosts that list both models.
Context window · side by side
Bar length is relative to the larger of the two windows (100% = max of this pair). This is not pricing.
R1 0528 has about 1.3× the context window of the other in this pair.
R1 0528 has 28% more context capacity (163K vs 128K tokens). R1 0528 is 75% cheaper on input.
Quick verdicts
Short takeaways — validate with your own workloads.
Long document processing
Use R1 0528. Its 163K context fits entire documents without chunking (vs 128K).
RAG / high-volume retrieval
Use R1 0528. Input tokens are 75% cheaper — critical when sending large retrieved contexts.
Full specs
Context, output, capabilities, and dates. Green highlights the favorable value where we compute a winner.
| Spec | R1 0528 | Sonar |
|---|---|---|
| Context window | 163,840 tokens (163K) | 128,000 tokens (128K) |
| Max output tokens | 163,840 tokens (163K) | N/A |
| Speed tier | Deep | Balanced |
| Vision | No | Yes |
| Function calling | Yes | No |
| Extended thinking | Yes | No |
| Prompt caching | Yes | No |
| Batch API | No | No |
| Release date | May 2025 | Jan 2025 |
Pricing matrix
Dollar rates only: hosts that list both models, per 1M tokens. For how much text fits, use the context section above — not this table.
| Provider | R1 0528 in | R1 0528 out | Sonar in | Sonar out |
|---|---|---|---|---|
| Deepinfra | $0.500/M | $2.15/M | — | — |
| Fireworks | $3.00/M | $8.00/M | — | — |
| Hyperbolic | $0.250/M | $0.250/M | — | — |
| Lambda | $0.200/M | $0.600/M | — | — |
| Nebius | $0.800/M | $2.40/M | — | — |
| Novita | $0.700/M | $2.50/M | — | — |
| Openrouter | $0.500/M | $2.15/M | — | — |
| Perplexity | — | — | $1.00/M | $1.00/M |
Frequently asked questions
Powered by Mem0
Use a smaller model.
Get better results.
Mem0 gives your AI long-term memory so you stop re-sending context on every call. That means you can use a smaller, faster, cheaper model — and still get better answers.
Example: a multi-turn chat session
80% less to send — works with any model