Mixtral Turbo Base: pricing, context window and benchmarks
Mixtral Turbo Base is a open-weight model from Together AI. It offers a 32,768-token context window, costs $2.53 per million input tokens and $10.80 per million output tokens, and averages 69.7 across the index's benchmark suite.
Pricing
Mixtral Turbo Base costs $2.53 per million input tokens and $10.80 per million output tokens. A blended 3:1 input-to-output mix works out to about $4.60 per million tokens. Cached input is listed at $0.32 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $2.53 | per 1M tokens |
| Output | $10.80 | per 1M tokens |
| Cached input | $0.32 | per 1M tokens |
| Blended 3:1 | $4.60 | estimate per 1M tokens |
| 1M in + 1M out | $13.33 | 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 | 70.7 | 390 | 35th | |
| GPQA Diamond | 63.7 | 380 | 37th | |
| SWE-bench Verified | 57.9 | 356 | 41th | |
| HumanEval | 72.9 | 374 | 38th | |
| MATH-500 | 74.0 | 347 | 42th | |
| AIME 2025 | 61.5 | 378 | 37th | |
| LiveCodeBench | 62.1 | 347 | 42th | |
| IFEval | 79.5 | 343 | 43th | |
| TruthfulQA | 74.2 | 349 | 42th | |
| HellaSwag | 82.8 | 327 | 46th | |
| MMMU | 67.9 | 344 | 43th | |
| Arena Elo | 1355.8 | 354 | 41th |
What Mixtral Turbo Base is good for
Mixtral Turbo 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 $2.53 per million input tokens (hosted list price) and a 32,768-token context window, it is a practical option for teams that want more control than a closed API allows.
Strengths
- Input price of $2.53 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
- Average benchmark score is below the index median; test it on your own task.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Phi-3.5 Instruct | Microsoft | 16384 | $2.67 | 70.1 |
| Reka Core Mini | Reka AI | 8192 | $7.48 | 70.1 |
| Firesearch Max | Fireworks AI | 8192 | $7.80 | 69.9 |
| Arctic Chat | Snowflake | 16384 | $2.70 | 69.8 |
| Mixtral Turbo Base | Together AI | 32768 | $2.53 | 69.7 |
| Gemini 3 Flash Nano | Google DeepMind | 32768 | $7.56 | 69.4 |
FAQ
How much does Mixtral Turbo Base cost?
Mixtral Turbo Base is listed at $2.53 per million input tokens and $10.80 per million output tokens. Cached input is $0.32 per million tokens. A blended 3:1 input-to-output mix is about $4.60 per million tokens.
What is the context window of Mixtral Turbo Base?
Mixtral Turbo Base has a context window of 32,768 tokens and can emit up to 8,192 output tokens per request in this index.
Is Mixtral Turbo Base open weights?
Yes. Mixtral Turbo Base is listed as an open-weight model under a open weights licence.
How does Mixtral Turbo Base score on benchmarks?
Mixtral Turbo Base averages 69.7 across the scored benchmarks in this index, ranking #367 of 600 models. Its strongest relative result is Arena Elo at 1355.8.
What is Mixtral Turbo Base best for?
Mixtral Turbo 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 $2.53 per million input tokens (hosted list price) and a 32,768-token context window, it is a practical option for teams that want more control than a closed API allows.