Primary use stated or inferred cautiously from official documentation.
R
RICURSIVE INTELLIGENCE / ricursiveAI/Ripple-65B-Preview-DFlash2
Ripple DFlash2
Auxiliary BF16 model with 4,037,752,064 parameters and five layers to accelerate Ripple through speculative decoding. It proposes token blocks from the target model’s internal states, which the target then verifies. It requires Ripple-65B Preview or its NVFP4 variant and does not work as an independent assistant.
01 / SUMMARY
The essential
Maximum input capacity when published by the source.
Documented maximum generation limit.
Verified availability channels.
Primary functional family and verified specialties.
Declared terms for API access, model weights, or self-hosting.
02 / TECHNICAL SHEET
Limits and integration
| API ID | ricursiveAI/Ripple-65B-Preview-DFlash2 |
|---|---|
| Model type | Inference acceleration |
| Access model | Open weights |
| License | Apache 2.0 |
| Deployment | Self-hosted |
| Release | 04 NOV 2026 |
| Knowledge cutoff | Not published |
| Entrance | Target model internal states |
| Exit | Candidate tokens |
| Context window | 131,072 configured positions |
| maximum output | Blocks of 16 candidate tokens |
| Reasoning | Not published |
| Published tools | |
| Structured Outings | Not published |
| Batch processing | Not published |
| Prompt cache | Not published |
| Fine-tuning | Not published |
| Verified platforms | Hugging Face · PyTorch |
Research and evaluation with your own data, preserving the exact weight revision, configuration and dependencies. Compare accuracy, memory and latency before integrating the model into an application.
The weights exclude embeddings and the output head shared with the target. The configured 131,072 positions are not an independent chat window. Export validation used eight H100 GPUs and a particular target revision; it does not establish compatibility with any server or version.
03 / CAPABILITIES
What it can do
Orientation
Research and evaluation with your own data, preserving the exact weight revision, configuration and dependencies. Compare accuracy, memory and latency before integrating the model into an application.
Context
131,072 configured positions · Blocks of 16 candidate tokens
Tools and integration
Not published
Access
Hugging Face · downloadable checkpoint
Input modalities
Target model internal states
04 / PRICES
Documented cost
| Concept | Worth | Unit/condition |
|---|---|---|
| Standard input | No published official API rate | Standard API rate |
| Cached input | Not published | Reading reused prefixes |
| Cache write or storage | Not published | The condition varies by provider |
| Standard output | Self-hosted infrastructure cost | May include reasoning tokens |
| Batch input | Not published | Asynchronous processing |
| Batch output | Not published | Asynchronous processing |
The weights require your own infrastructure. Cost depends on memory, precision, sequence length and load. Review the license and runtime code before deployment. A published performance figure does not guarantee the same cost or speed on another machine.
i Prices change and may depend on level, region or context length. Check the source before making a decision.
05 / EVALUATIONS
How to read the results
“
No public acceleration figure has been verified for this exact artifact. Speed depends on candidate acceptance, load and runtime. Check that speculative sampling preserves the target model’s distribution. Ripple’s reasoning or coding scores are not transferred to this drafter.
i No verified numeric score for the exact version.
Inferama only highlights a “best result” when the metric, test set, configuration, and date allow for an equivalent comparison. The supplier's figures are presented as claims from its own source.
06 / VERSIONS
Chronology
Ripple DFlash2
Version added to Inferama's verified catalog.
07 / SOURCES