Leaderboard / Essential AI (Rnj)
Rnj-1.5 Instruct
Essential AI (Rnj) · Rnj · released 2026-04-22 · data checked 2026-11-01
Rnj-1.5 Instruct is a 8.3B-parameter dense model from Essential AI (Rnj) (United States), released 2026-04-22 under the Apache-2.0 license. It doesn't have enough reported benchmarks for an index score yet. It needs about 6.8 GB of memory at 4-bit with an 8K-token context.
Coding-focused 8B model; results are published as a chart.
Small LM Index
—
Needs 2 core benchmarks (has 0)
Parameters
8.3B
Dense
Price per 1M tokens
—
No hosted API on OpenRouter; self-host
Context window
164K
tokens
Benchmarks
Each bar shows this model's score. The tick marks the best small model on that benchmark.
No benchmark results for this release yet. Know of one? Send it.
Specifications
- Lab
- Essential AI (Rnj) (US · United States)
- Architecture
- Dense transformer
- Total parameters
- 8.3B
- Active parameters
- All (dense)
- Size class
- Small
- Context window
- 164K tokens
- Input modalities
- text
- Reasoning
- Instruct (no thinking)
- License
- Apache-2.0
- Openness
- Open weights
- Memory to run
- ~6.8 GB at 4-bit, 8K context (more sizes)
- Hugging Face
- 1.2K downloads (30 days) · 21 likes
Compare with
- Rnj-1.5 Instruct vs DeepSeek-R1-0528-Qwen3-8B (8.2B, index 53.1)
- Rnj-1.5 Instruct vs Qwen3 8B (8.2B, index 48.5)
- Rnj-1.5 Instruct vs Granite 4.2 8B (8B, index 63.5)
Sources
Something wrong here? Report it.
Run it locally
No GGUF build from a well-known publisher yet.
vLLM (GPU server)
vllm serve EssentialAI/rnj-1.5-instructFull-precision weights, for a server GPU.
Memory needed: weights, context cache and runtime buffers.
| Quantization | 4K context | 32K context | 164K context |
|---|---|---|---|
| 16-bit F16 | 18 GB | 19 GB | 25 GB |
| 8-bit Q8_0 | 10 GB | 12 GB | 18 GB |
| 4-bit Q4_K_M | 6.2 GB | 7.9 GB | 14 GB |