Pythia 8B: pricing, context window and benchmarks
Pythia 8B is a small open-weight model from EleutherAI. It offers a 32,768-token context window, costs $0.16 per million input tokens and $0.79 per million output tokens, and averages 41.4 across the index's benchmark suite.
Pricing
Pythia 8B costs $0.16 per million input tokens and $0.79 per million output tokens. A blended 3:1 input-to-output mix works out to about $0.32 per million tokens. Cached input is listed at $0.02 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $0.16 | per 1M tokens |
| Output | $0.79 | per 1M tokens |
| Cached input | $0.02 | per 1M tokens |
| Blended 3:1 | $0.32 | estimate per 1M tokens |
| 1M in + 1M out | $0.95 | 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 | 50.3 | 593 | 1th | |
| GPQA Diamond | 38.2 | 592 | 1th | |
| SWE-bench Verified | 27.3 | 588 | 2th | |
| HumanEval | 42.8 | 594 | 1th | |
| MATH-500 | 37.9 | 595 | 1th | |
| AIME 2025 | 29.8 | 584 | 3th | |
| LiveCodeBench | 24.1 | 594 | 1th | |
| IFEval | 53.9 | 593 | 1th | |
| TruthfulQA | 49.7 | 594 | 1th | |
| HellaSwag | 60.7 | 595 | 1th | |
| MMMU | 40.5 | 593 | 1th | |
| Arena Elo | 1130.8 | 592 | 1th |
What Pythia 8B is good for
Pythia 8B is a small model. That makes it a good fit for classification, extraction, routing, summarization and other high-volume jobs where latency and cost dominate. It is not the right pick for open-ended frontier reasoning, but at $0.16 per million input tokens it can handle a large share of production traffic.
Strengths
- Input price of $0.16 per million tokens is at or below the index median.
- Open weights make self-hosting, fine-tuning and data-residency review possible.
Trade-offs
- Average benchmark score is below the index median; test it on your own task.
- Less reliable on open-ended reasoning and complex multi-step instructions.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Molmo 1.5B | Allen Institute for AI | 8192 | $0.21 | 42.8 |
| Zephyr 7B | Hugging Face | 32768 | $0.23 | 42.5 |
| GPT-NeoX 7B | EleutherAI | 8192 | $0.20 | 42.4 |
| Nemotron Nano 1.5B | NVIDIA | 8192 | $0.10 | 41.5 |
| Pythia 8B | EleutherAI | 32768 | $0.16 | 41.4 |
| Pythia 7B | EleutherAI | 8192 | $0.10 | 40.7 |
FAQ
How much does Pythia 8B cost?
Pythia 8B is listed at $0.16 per million input tokens and $0.79 per million output tokens. Cached input is $0.020 per million tokens. A blended 3:1 input-to-output mix is about $0.32 per million tokens.
What is the context window of Pythia 8B?
Pythia 8B has a context window of 32,768 tokens and can emit up to 32,768 output tokens per request in this index.
Is Pythia 8B open weights?
Yes. Pythia 8B is listed as an open-weight model under a open weights licence.
How does Pythia 8B score on benchmarks?
Pythia 8B averages 41.4 across the scored benchmarks in this index, ranking #594 of 600 models. Its strongest relative result is Arena Elo at 1130.8.
What is Pythia 8B best for?
Pythia 8B is a small model. That makes it a good fit for classification, extraction, routing, summarization and other high-volume jobs where latency and cost dominate. It is not the right pick for open-ended frontier reasoning, but at $0.16 per million input tokens it can handle a large share of production traffic.