DeepSeek
deepseek-r1-distill-llama-8b
$0.025 / 1M tokens
cheapest at Nscale · $0.025 in · $0.025 out
Wins 1 of 3
Head to head · prices updated Oct 31, 2026
deepseek-r1-distill-llama-8b vs gpt-oss-20b
gpt-oss-20b costs 2.2× as much as deepseek-r1-distill-llama-8b at their cheapest providers (blended 3 input : 1 output tokens).
OpenAI
gpt-oss-20b
$0.055 / 1M tokens
cheapest at W&B Inference (CoreWeave) · $0.03 in · $0.13 out
Wins 1 of 3
- Lower pricedeepseek-r1-distill-llama-8b
- Longer contextTie
- More providersgpt-oss-20b
Verdict
deepseek-r1-distill-llama-8b is the better deal on price
gpt-oss-20b costs 2.2× as much as deepseek-r1-distill-llama-8b at their cheapest providers (blended 3 input : 1 output tokens).
Your monthly bill
What would each cost you?
deepseek-r1-distill-llama-8b saves you $0.26 a month ($3.12 a year) for this workload.
Side by side
| deepseek-r1-distill-llama-8b | gpt-oss-20b | |
|---|---|---|
| Cheapest blended price / 1M | $0.025 | $0.055 |
| Input / output per 1M | $0.025 / $0.025 | $0.03 / $0.13 |
| Accuracy (GPQA Diamond) | not measured | 60.8% (medium reasoning) |
| Max context | 131K | 131K |
| Providers we track | 2 | 10 |
Standard list prices at each model's cheapest provider we track; blended = 3 input : 1 output tokens. GPQA Diamond from Epoch AI (CC BY 4.0), best measured reasoning setting, named next to each score. Methodology.
Where to buy each one
gpt-oss-20b
- W&B Inference (CoreWeave)$0.055
- DeepInfra$0.0575
- Novita AI$0.0675
- SiliconFlow$0.075
- OVHcloud AI Endpoints$0.0754
Questions people ask
Is deepseek-r1-distill-llama-8b cheaper than gpt-oss-20b?
gpt-oss-20b costs 2.2× as much as deepseek-r1-distill-llama-8b at their cheapest providers (blended 3 input : 1 output tokens).
How much does deepseek-r1-distill-llama-8b cost per 1M tokens?
From $0.025 input and $0.025 output per 1M tokens at Nscale, the cheapest of 2 providers we track.
How much does gpt-oss-20b cost per 1M tokens?
From $0.03 input and $0.13 output per 1M tokens at W&B Inference (CoreWeave), the cheapest of 10 providers we track.
Swap a model