Llama 3 DeepInfra Base: pricing, context window and benchmarks
Llama 3 DeepInfra Base is a open-weight model from DeepInfra. It offers a 16,384-token context window, costs $2.95 per million input tokens and $15.78 per million output tokens, and averages 71.8 across the index's benchmark suite.
Pricing
Llama 3 DeepInfra Base costs $2.95 per million input tokens and $15.78 per million output tokens. A blended 3:1 input-to-output mix works out to about $6.16 per million tokens. Cached input is listed at $0.40 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $2.95 | per 1M tokens |
| Output | $15.78 | per 1M tokens |
| Cached input | $0.40 | per 1M tokens |
| Blended 3:1 | $6.16 | estimate per 1M tokens |
| 1M in + 1M out | $18.73 | estimate |
Benchmarks
Scores below are modelled estimates on a 0–100 scale. Use them to compare relative position, not as official vendor results.
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 75.6 | 316 | 47th | |
| GPQA Diamond | 64.5 | 371 | 38th | |
| SWE-bench Verified | 58.2 | 352 | 41th | |
| HumanEval | 76.2 | 326 | 46th | |
| MATH-500 | 75.9 | 302 | 50th | |
| AIME 2025 | 67.5 | 302 | 50th | |
| LiveCodeBench | 63.5 | 332 | 45th | |
| IFEval | 79.7 | 339 | 44th | |
| TruthfulQA | 77.0 | 282 | 53th | |
| HellaSwag | 83.8 | 287 | 52th | |
| MMMU | 67.6 | 351 | 42th | |
| Arena Elo | 1374.7 | 295 | 51th |
What Llama 3 DeepInfra Base is good for
Llama 3 DeepInfra Base is an open-weight model, so it can be self-hosted or run through a third-party inference provider. That matters for data residency, fine-tuning and long-run cost control. At $2.95 per million input tokens (hosted list price) and a 16,384-token context window, it is a practical option for teams that want more control than a closed API allows.
Strengths
- Input price of $2.95 per million tokens is at or below the index median.
- Open weights make self-hosting, fine-tuning and data-residency review possible.
- Accepts more than text, including image input where the provider supports it.
Trade-offs
- Context window of 16,384 tokens is small for long-document work.
- Average benchmark score is below the index median; test it on your own task.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Llama 3 DeepInfra Base | DeepInfra | 16384 | $2.95 | 71.8 |
| DBRX Chat | Databricks | 65536 | $3.07 | 71.7 |
| Fusion Turbo | OpenRouter | 8192 | $8.55 | 71.7 |
| Mixtral Turbo Chat | Together AI | 32768 | $3.02 | 71.7 |
| OLMo 2 Instruct | Allen Institute for AI | 32768 | $3.03 | 71.7 |
| Qwen3-Max Nano | Alibaba Cloud | 65536 | $8.31 | 71.7 |
FAQ
How much does Llama 3 DeepInfra Base cost?
Llama 3 DeepInfra Base is listed at $2.95 per million input tokens and $15.78 per million output tokens. Cached input is $0.40 per million tokens. A blended 3:1 input-to-output mix is about $6.16 per million tokens.
What is the context window of Llama 3 DeepInfra Base?
Llama 3 DeepInfra Base has a context window of 16,384 tokens and can emit up to 4,096 output tokens per request in this index.
Is Llama 3 DeepInfra Base open weights?
Yes. Llama 3 DeepInfra Base is listed as an open-weight model under a open weights licence.
How does Llama 3 DeepInfra Base score on benchmarks?
Llama 3 DeepInfra Base averages 71.8 across the scored benchmarks in this index, ranking #331 of 600 models. Its strongest relative result is Arena Elo at 1374.7.
What is Llama 3 DeepInfra Base best for?
Llama 3 DeepInfra Base is an open-weight model, so it can be self-hosted or run through a third-party inference provider. That matters for data residency, fine-tuning and long-run cost control. At $2.95 per million input tokens (hosted list price) and a 16,384-token context window, it is a practical option for teams that want more control than a closed API allows.