Reka Core: pricing, context window and benchmarks
Reka Core is a multimodal model from Reka AI. It offers a 32,768-token context window, costs $12.17 per million input tokens and $55.77 per million output tokens, and averages 77.4 across the index's benchmark suite.
Pricing
Reka Core costs $12.17 per million input tokens and $55.77 per million output tokens. A blended 3:1 input-to-output mix works out to about $23.07 per million tokens. Cached input is listed at $1.24 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $12.17 | per 1M tokens |
| Output | $55.77 | per 1M tokens |
| Cached input | $1.24 | per 1M tokens |
| Blended 3:1 | $23.07 | estimate per 1M tokens |
| 1M in + 1M out | $67.94 | 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 | 80.0 | 185 | 69th | |
| GPQA Diamond | 74.8 | 148 | 75th | |
| SWE-bench Verified | 63.8 | 238 | 60th | |
| HumanEval | 82.0 | 187 | 69th | |
| MATH-500 | 83.8 | 130 | 78th | |
| AIME 2025 | 72.2 | 217 | 64th | |
| LiveCodeBench | 69.5 | 202 | 66th | |
| IFEval | 85.7 | 169 | 72th | |
| TruthfulQA | 79.0 | 221 | 63th | |
| HellaSwag | 87.1 | 175 | 71th | |
| MMMU | 73.4 | 209 | 65th | |
| Arena Elo | 1422.8 | 136 | 77th |
What Reka Core is good for
Reka Core is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $12.17 per million input tokens, it sits in the mid-range of the index, and the 32,768-token context window is enough for large files and multi-file diffs.
Strengths
- Accepts more than text, including image input where the provider supports it.
- Coding benchmarks are a relative strength within its tier.
Trade-offs
- Proprietary licence: you cannot download the weights or self-host the model.
- Premium input pricing at $12.17 per million tokens.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| ERNIE 4.0 Pro | Baidu | 200000 | $12.58 | 77.5 |
| Hunyuan-Turbo Reasoning | Tencent | 128000 | $11.89 | 77.5 |
| Hunyuan-Turbo Turbo | Tencent | 65536 | $11.60 | 77.4 |
| Kimi K1.5 Base | Moonshot AI | 65536 | $4.34 | 77.4 |
| Qwen3 Base | Alibaba Cloud | 65536 | $4.33 | 77.4 |
| Reka Core | Reka AI | 32768 | $12.17 | 77.4 |
FAQ
How much does Reka Core cost?
Reka Core is listed at $12.17 per million input tokens and $55.77 per million output tokens. Cached input is $1.24 per million tokens. A blended 3:1 input-to-output mix is about $23.07 per million tokens.
What is the context window of Reka Core?
Reka Core has a context window of 32,768 tokens and can emit up to 16,384 output tokens per request in this index.
Is Reka Core open weights?
No. Reka Core is a proprietary model with a proprietary licence, so you cannot download the weights.
How does Reka Core score on benchmarks?
Reka Core averages 77.4 across the scored benchmarks in this index, ranking #190 of 600 models. Its strongest relative result is Arena Elo at 1422.8.
What is Reka Core best for?
Reka Core is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $12.17 per million input tokens, it sits in the mid-range of the index, and the 32,768-token context window is enough for large files and multi-file diffs.