R1
DeepSeek released 2025-01-20
$1.15 per million tokens, balanced
DeepSeek V4 Flash 0731 is both better and cheaper.
It scores +33.2 higher and costs 91% less ($0.105/M against $1.15/M) on this workload — and it does everything this model does.
KAT-Coder-Pro V2 is cheaper still (54% less) but drops no extended reasoning.
Every price dimension
| Input | $0.700/M |
|---|---|
| Output | $2.50/M |
Independent scores
| Intelligence | 18.6 |
|---|---|
| Coding | 24.6 |
| Agentic | 3.1 |
| LMArena Elo | 1372.7default |
Sources are listed separately rather than averaged. Across the 62 models both have scored they correlate at r = 0.806 — close agreement overall, but four models rank very differently between them, and a blended score would hide exactly those.
Capability
| Context window | 64K |
|---|---|
| Max output | 16K |
| Input modes | text |
| Tool use | yes |
| Reasoning | always on |
| Knowledge cutoff | 2024-07-31 |
Provenance
| Price source | openrouter.ai |
|---|---|
| Fetched | 2026-08-24 |
| Quality data | verified |
DeepSeek R1 is here: Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active in an inference pass....