StableLM 2: pricing, context window and benchmarks
StableLM 2 is a open-weight model from Stability AI. It offers a 8,192-token context window, costs $1.91 per million input tokens and $9.65 per million output tokens, and averages 65.0 across the index's benchmark suite.
Pricing
StableLM 2 costs $1.91 per million input tokens and $9.65 per million output tokens. A blended 3:1 input-to-output mix works out to about $3.84 per million tokens. Cached input is listed at $0.27 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $1.91 | per 1M tokens |
| Output | $9.65 | per 1M tokens |
| Cached input | $0.27 | per 1M tokens |
| Blended 3:1 | $3.84 | estimate per 1M tokens |
| 1M in + 1M out | $11.56 | 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 | 71.4 | 387 | 36th | |
| GPQA Diamond | 57.9 | 429 | 29th | |
| SWE-bench Verified | 52.0 | 407 | 32th | |
| HumanEval | 70.0 | 394 | 34th | |
| MATH-500 | 68.3 | 397 | 34th | |
| AIME 2025 | 56.5 | 408 | 32th | |
| LiveCodeBench | 55.2 | 402 | 33th | |
| IFEval | 73.2 | 413 | 31th | |
| TruthfulQA | 68.7 | 408 | 32th | |
| HellaSwag | 77.2 | 413 | 31th | |
| MMMU | 64.2 | 387 | 36th | |
| Arena Elo | 1294.3 | 426 | 29th |
What StableLM 2 is good for
StableLM 2 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 $1.91 per million input tokens (hosted list price) and a 8,192-token context window, it is a practical option for teams that want more control than a closed API allows.
Strengths
- Input price of $1.91 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 8,192 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 |
|---|---|---|---|---|
| StableLM 2 Instruct | Stability AI | 16384 | $1.93 | 65.2 |
| Reka Core Nano | Reka AI | 16384 | $5.45 | 65.1 |
| Molmo Base | Allen Institute for AI | 4096 | $2.02 | 65.0 |
| Nemotron Nano Base | NVIDIA | 8192 | $1.85 | 65.0 |
| Nova Lite Mini | Amazon | 16384 | $5.43 | 65.0 |
| StableLM 2 | Stability AI | 8192 | $1.91 | 65.0 |
FAQ
How much does StableLM 2 cost?
StableLM 2 is listed at $1.91 per million input tokens and $9.65 per million output tokens. Cached input is $0.27 per million tokens. A blended 3:1 input-to-output mix is about $3.84 per million tokens.
What is the context window of StableLM 2?
StableLM 2 has a context window of 8,192 tokens and can emit up to 8,192 output tokens per request in this index.
Is StableLM 2 open weights?
Yes. StableLM 2 is listed as an open-weight model under a open weights licence.
How does StableLM 2 score on benchmarks?
StableLM 2 averages 65.0 across the scored benchmarks in this index, ranking #406 of 600 models. Its strongest relative result is Arena Elo at 1294.3.
What is StableLM 2 best for?
StableLM 2 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 $1.91 per million input tokens (hosted list price) and a 8,192-token context window, it is a practical option for teams that want more control than a closed API allows.