R1 Distill Llama 70B
by DeepSeek · deepseek/deepseek-r1-distill-llama-70b · #159 cheapest of 325 paid models
Prices updated Jul 22, 2026, 2:52 AM UTC · refreshed hourly
DeepSeek R1 Distill Llama 70B is a distilled large language model based on [Llama-3.3-70B-Instruct](/meta-llama/llama-3.3-70b-instruct), using outputs from [DeepSeek R1](/deepseek/deepseek-r1). The model combines advanced distillation techniques to achieve high performance across...
Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5GPT-5.6 Sol
Pricing
Input
$0.80
per 1M tokens
Output
$0.80
per 1M tokens
Blended (3:1)
$0.80
3 input : 1 output
Cache read
—
per 1M cached tokens
Cache write
—
per 1M tokens
Cache & batch economics
Effective prices for R1 Distill Llama 70B (no cache-read rate published — hits billed as normal input).
| Cache hit rate | Effective blended $/1M | Example request* | vs no cache |
|---|---|---|---|
| 0% | $0.80 | $0.0824 | — |
| 50% | $0.80 | $0.0824 | −0% |
| 90% | $0.80 | $0.0824 | −0% |
*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates (available for OpenAI / Anthropic / Google — not flagged for this creator). Confirm on the vendor’s pricing page.
What a request costs
| Workload | Input tokens | Output tokens | Cost / request | Cost / 1K requests |
|---|---|---|---|---|
| Short chat message | 500 | 300 | $0.00064 | $0.64 |
| Document summary | 8,000 | 1,000 | $0.0072 | $7.20 |
| Codebase question (RAG) | 30,000 | 2,000 | $0.0256 | $25.60 |
| Long-context analysis | 150,000 | 5,000 | $0.124 | $124.00 |
Performance
Intelligence Index
9.9
AA composite quality
Coding Index
—
Math Index
53.7
Agentic Index
—
Output speed
34 tok/s
median
Time to first token
0.59s
median
Time to first answer
60.07s
after reasoning tokens
Artificial Analysis benchmarks
| Benchmark | Score |
|---|---|
| MMLU-Pro | 79.5% |
| GPQA Diamond | 40.2% |
| Humanity's Last Exam | 6.1% |
| LiveCodeBench | 26.6% |
| SciCode | 31.3% |
| MATH-500 | 93.5% |
| AIME | 67.0% |
| AIME 2025 | 53.7% |
| IFBench | 27.6% |
| AA-LCR (Long Context Reasoning) | 11.0% |
| Terminal-Bench Hard | 1.5% |
| τ²-Bench (Telecom) | 21.9% |
Specs
Context window
8K
tokens
Max output
8K
tokens
Input modalities
text
Tokenizer
Llama3