Best DeepSeek R1 Alternatives
DeepSeek R1 by DeepSeek is a reasoning model priced at $0.55/2.19 per 1M tokens (in/out). It's already affordable, but you might want different strengths or features.
DeepSeek R1
DeepSeekReasoningInput
$0.55/1M
Output
$2.19/1M
Context
128K
Max Output
64K
Why Switch from DeepSeek R1?
Top Alternatives
27% cheaper, higher benchmark scores, adds vision, tool-use.
Input
$0.4/1M
27% cheaper
Output
$1.6/1M
27% cheaper
Context
200K
Max Output
100K
45% cheaper, higher benchmark scores, open-source and self-hostable.
Input
$0.3/1M
45% cheaper
Output
$1.2/1M
45% cheaper
Context
200K
Max Output
128K
Higher benchmark scores, adds vision, tool-use.
Input
$1.1/1M
100% more
Output
$4.4/1M
101% more
Context
200K
Max Output
100K
Dramatically cheaper (85% less), comparable performance, open-source and self-hostable.
Input
$0.14/1M
75% cheaper
Output
$0.28/1M
87% cheaper
Context
164K
Max Output
16K
Dramatically cheaper (58% less), higher benchmark scores, 1M context (8x more).
Input
$0.31/1M
44% cheaper
Output
$0.85/1M
61% cheaper
Context
1M
Max Output
32K
Higher benchmark scores, adds vision.
Input
$0.6/1M
9% more
Output
$2.2/1M
0% more
Context
200K
Max Output
128K
Comparable performance, 1M context (8x more), adds vision, tool-use.
Input
$0.5/1M
9% cheaper
Output
$3/1M
37% more
Context
1M
Max Output
66K
Dramatically cheaper (73% less), comparable performance, 1M context (8x more).
Input
$0.15/1M
73% cheaper
Output
$0.6/1M
73% cheaper
Context
1M
Max Output
66K
Full Comparison Table
| Model | Input $/1M | Output $/1M | Context | MMLU-Pro | HumanEval | Score |
|---|---|---|---|---|---|---|
| o3OpenAI | $0.4027% cheaper | $1.6027% cheaper | 200K | 87%+3.0% | 94.5%+2.5% | 88 |
| MiniMax M2.5MiniMax | $0.3045% cheaper | $1.2045% cheaper | 200K | 82%-2.0% | 90%-2.0% | 85 |
| o4-miniOpenAI | $1.10100% more | $4.40101% more | 200K | 85%+1.0% | 93.5%+1.5% | 78 |
| DeepSeek V3DeepSeek | $0.1475% cheaper | $0.2887% cheaper | 164K | 78%-6.0% | 89%-3.0% | 75 |
| Llama 4 MaverickMeta | $0.3144% cheaper | $0.8561% cheaper | 1M | 80.5%-3.5% | 90.2%-1.8% | 70 |
| GLM-4.7Zhipu AI | $0.609% more | $2.200% more | 200K | 84.3%+0.3% | — | 68 |
| Gemini 3 FlashGoogle | $0.509% cheaper | $3.0037% more | 1M | 78%-6.0% | 90%-2.0% | 63 |
| Gemini 2.5 FlashGoogle | $0.1573% cheaper | $0.6073% cheaper | 1M | 76%-8.0% | 89.5%-2.5% | 63 |
| Mistral Large 3Mistral | $2.00264% more | $5.00128% more | 128K | 83%-1.0% | 91%-1.0% | 63 |
| Mistral Medium 3Mistral | $0.4027% cheaper | $2.009% cheaper | 128K | 76%-8.0% | 87%-5.0% | 60 |
| GLM-5Zhipu AI | $1.0082% more | $3.2046% more | 200K | 70.4%-13.6% | 91%-1.0% | 56 |
| Claude Sonnet 4.6Anthropic | $3.00445% more | $15.00585% more | 200K | 86%+2.0% | 94%+2.0% | 53 |
| GPT-5.3 CodexOpenAI | $2.00264% more | $16.00631% more | 200K | 90%+6.0% | 96.5%+4.5% | 53 |
| GPT-5.2 CodexOpenAI | $1.75218% more | $14.00539% more | 200K | 89%+5.0% | 95.5%+3.5% | 53 |
| Llama 4 ScoutMeta | $0.1867% cheaper | $0.6371% cheaper | 10M | 74.2%-9.8% | 86%-6.0% | 53 |
| Claude Haiku 4.5Anthropic | $0.8045% more | $4.0083% more | 200K | 69.4%-14.6% | 88.1%-3.9% | 50 |
| GPT-5OpenAI | $1.25127% more | $10.00357% more | 128K | 88.5%+4.5% | 95%+3.0% | 50 |
| GPT-4o MiniOpenAI | $0.1573% cheaper | $0.6073% cheaper | 128K | 68%-16.0% | 87.2%-4.8% | 50 |
| Gemini 3.1 ProGoogle | $2.00264% more | $12.00448% more | 1M | 91%+7.0% | 95%+3.0% | 50 |
| Gemini 3 ProGoogle | $2.00264% more | $12.00448% more | 1M | 89.8%+5.8% | 94%+2.0% | 50 |
| Gemini 2.5 ProGoogle | $1.25127% more | $10.00357% more | 1M | 87.5%+3.5% | 93.5%+1.5% | 50 |
| Grok 4xAI | $3.00445% more | $15.00585% more | 128K | 86%+2.0% | 93%+1.0% | 50 |
| Claude Sonnet 4.5Anthropic | $3.00445% more | $15.00585% more | 200K | 84.5%+0.5% | 93%+1.0% | 46 |
| Claude Opus 4.6Anthropic | $5.00809% more | $25.001042% more | 200K | 89.5%+5.5% | 95%+3.0% | 43 |
| GPT-4oOpenAI | $2.50355% more | $10.00357% more | 128K | 80.5%-3.5% | 91%-1.0% | 38 |