Seed-OSS 7B: pricing, context window and benchmarks
Seed-OSS 7B is a small open-weight model from ByteDance. It offers a 8,192-token context window, costs $0.85 per million input tokens and $3.81 per million output tokens, and averages 55.5 across the index's benchmark suite.
Pricing
Seed-OSS 7B costs $0.85 per million input tokens and $3.81 per million output tokens. A blended 3:1 input-to-output mix works out to about $1.59 per million tokens. Cached input is listed at $0.10 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $0.85 | per 1M tokens |
| Output | $3.81 | per 1M tokens |
| Cached input | $0.10 | per 1M tokens |
| Blended 3:1 | $1.59 | estimate per 1M tokens |
| 1M in + 1M out | $4.66 | 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 | 61.1 | 512 | 15th | |
| GPQA Diamond | 51.7 | 504 | 16th | |
| SWE-bench Verified | 43.6 | 484 | 19th | |
| HumanEval | 59.9 | 506 | 16th | |
| MATH-500 | 54.4 | 525 | 13th | |
| AIME 2025 | 43.1 | 522 | 13th | |
| LiveCodeBench | 44.8 | 501 | 17th | |
| IFEval | 65.3 | 529 | 12th | |
| TruthfulQA | 61.9 | 508 | 15th | |
| HellaSwag | 72.4 | 489 | 19th | |
| MMMU | 51.8 | 534 | 11th | |
| Arena Elo | 1239.5 | 515 | 14th |
What Seed-OSS 7B is good for
Seed-OSS 7B 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 $0.85 per million input tokens it can handle a large share of production traffic.
Strengths
- Input price of $0.85 per million tokens is at or below the index median.
- Open weights make self-hosting, fine-tuning and data-residency review possible.
Trade-offs
- Context window of 8,192 tokens is small for long-document work.
- 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 |
|---|---|---|---|---|
| Firesearch Nano | Fireworks AI | 8192 | $2.49 | 55.6 |
| Llama 3 DeepInfra 7B | DeepInfra | 16384 | $0.95 | 55.6 |
| Llama 3 Hyperbolic 7B | Hyperbolic | 4096 | $0.94 | 55.6 |
| Llama 3 Novita 8B | Novita AI | 16384 | $0.99 | 55.5 |
| Seed-OSS 7B | ByteDance | 8192 | $0.85 | 55.5 |
| Mistral Large 3 1.5B | Mistral AI | 32768 | $0.82 | 55.4 |
FAQ
How much does Seed-OSS 7B cost?
Seed-OSS 7B is listed at $0.85 per million input tokens and $3.81 per million output tokens. Cached input is $0.10 per million tokens. A blended 3:1 input-to-output mix is about $1.59 per million tokens.
What is the context window of Seed-OSS 7B?
Seed-OSS 7B has a context window of 8,192 tokens and can emit up to 4,096 output tokens per request in this index.
Is Seed-OSS 7B open weights?
Yes. Seed-OSS 7B is listed as an open-weight model under a open weights licence.
How does Seed-OSS 7B score on benchmarks?
Seed-OSS 7B averages 55.5 across the scored benchmarks in this index, ranking #515 of 600 models. Its strongest relative result is Arena Elo at 1239.5.
What is Seed-OSS 7B best for?
Seed-OSS 7B 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 $0.85 per million input tokens it can handle a large share of production traffic.