Mixtral Turbo Instruct: pricing, context window and benchmarks
Mixtral Turbo Instruct is a open-weight model from Together AI. It offers a 32,768-token context window, costs $3.16 per million input tokens and $13.04 per million output tokens, and averages 71.9 across the index's benchmark suite.
Pricing
Mixtral Turbo Instruct costs $3.16 per million input tokens and $13.04 per million output tokens. A blended 3:1 input-to-output mix works out to about $5.63 per million tokens. Cached input is listed at $0.41 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $3.16 | per 1M tokens |
| Output | $13.04 | per 1M tokens |
| Cached input | $0.41 | per 1M tokens |
| Blended 3:1 | $5.63 | estimate per 1M tokens |
| 1M in + 1M out | $16.20 | 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 | 72.9 | 369 | 39th | |
| GPQA Diamond | 69.7 | 276 | 54th | |
| SWE-bench Verified | 60.8 | 309 | 49th | |
| HumanEval | 74.2 | 355 | 41th | |
| MATH-500 | 75.9 | 303 | 50th | |
| AIME 2025 | 67.5 | 304 | 49th | |
| LiveCodeBench | 61.4 | 358 | 40th | |
| IFEval | 82.1 | 275 | 54th | |
| TruthfulQA | 77.2 | 276 | 54th | |
| HellaSwag | 82.5 | 336 | 44th | |
| MMMU | 66.6 | 365 | 39th | |
| Arena Elo | 1356.6 | 351 | 42th |
What Mixtral Turbo Instruct is good for
Mixtral Turbo Instruct 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 $3.16 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 $3.16 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 |
|---|---|---|---|---|
| Llama 3 Replicate Base | Replicate | 65536 | $3.15 | 72.1 |
| Claude Sonnet 4.5 Nano | Anthropic | 8192 | $8.93 | 72.0 |
| Mixtral Groq Chat | Groq | 8192 | $3.09 | 72.0 |
| o4 Nano | OpenAI | 65536 | $8.61 | 72.0 |
| Mixtral Turbo Instruct | Together AI | 32768 | $3.16 | 71.9 |
| Llama 3 DeepInfra Base | DeepInfra | 16384 | $2.95 | 71.8 |
FAQ
How much does Mixtral Turbo Instruct cost?
Mixtral Turbo Instruct is listed at $3.16 per million input tokens and $13.04 per million output tokens. Cached input is $0.41 per million tokens. A blended 3:1 input-to-output mix is about $5.63 per million tokens.
What is the context window of Mixtral Turbo Instruct?
Mixtral Turbo Instruct has a context window of 32,768 tokens and can emit up to 32,768 output tokens per request in this index.
Is Mixtral Turbo Instruct open weights?
Yes. Mixtral Turbo Instruct is listed as an open-weight model under a open weights licence.
How does Mixtral Turbo Instruct score on benchmarks?
Mixtral Turbo Instruct averages 71.9 across the scored benchmarks in this index, ranking #330 of 600 models. Its strongest relative result is Arena Elo at 1356.6.
What is Mixtral Turbo Instruct best for?
Mixtral Turbo Instruct 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 $3.16 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.