Update README.md
Browse files
README.md
CHANGED
|
@@ -13,4 +13,64 @@ tags:
|
|
| 13 |
|
| 14 |
# **Jan-v1-AIO-GGUF**
|
| 15 |
|
| 16 |
-
> Jan-v1-4B is a 4-billion-parameter language model built on the Qwen3-4B-thinking architecture, meticulously fine-tuned for agentic reasoning, problem-solving, and tool utilization with support for web search tasks and large context lengths up to 256,000 tokens. Achieving 91.1% accuracy on the SimpleQA benchmark, Jan-v1-4B excels at factual question answering and conversation while running efficiently on local hardware for enhanced privacy and offline use, making it a strong choice for advanced Q&A, reasoning, and integration with the Jan desktop application or compatible inference engines. Jan-v1-edge is a lightweight agentic model built for fast, reliable on-device execution. As the second release in the Jan Family, it is distilled from the larger Jan-v1 model, preserving strong reasoning and problem-solving ability in a smaller footprint suitable for resource-constrained environments. Jan-v1-edge was developed through a two-phase post-training process. The first phase, Supervised Fine-Tuning (SFT), transferred core capabilities from the Jan-v1 teacher model to the smaller student. The second phase, Reinforcement Learning with Verifiable Rewards (RLVR) —the same method used in Jan-v1 and Lucy—further optimized reasoning efficiency, tool use, and correctness. This staged approach delivers reliable results on complex, interactive workloads.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
|
| 14 |
# **Jan-v1-AIO-GGUF**
|
| 15 |
|
| 16 |
+
> Jan-v1-4B is a 4-billion-parameter language model built on the Qwen3-4B-thinking architecture, meticulously fine-tuned for agentic reasoning, problem-solving, and tool utilization with support for web search tasks and large context lengths up to 256,000 tokens. Achieving 91.1% accuracy on the SimpleQA benchmark, Jan-v1-4B excels at factual question answering and conversation while running efficiently on local hardware for enhanced privacy and offline use, making it a strong choice for advanced Q&A, reasoning, and integration with the Jan desktop application or compatible inference engines. Jan-v1-edge is a lightweight agentic model built for fast, reliable on-device execution. As the second release in the Jan Family, it is distilled from the larger Jan-v1 model, preserving strong reasoning and problem-solving ability in a smaller footprint suitable for resource-constrained environments. Jan-v1-edge was developed through a two-phase post-training process. The first phase, Supervised Fine-Tuning (SFT), transferred core capabilities from the Jan-v1 teacher model to the smaller student. The second phase, Reinforcement Learning with Verifiable Rewards (RLVR) —the same method used in Jan-v1 and Lucy—further optimized reasoning efficiency, tool use, and correctness. This staged approach delivers reliable results on complex, interactive workloads.
|
| 17 |
+
|
| 18 |
+
## Jan-v1 GGUF Models
|
| 19 |
+
|
| 20 |
+
| Model Name | Hugging Face Link |
|
| 21 |
+
|---------------|-------------------|
|
| 22 |
+
| **Jan-v1-edge-GGUF** | [🔗 Link](https://huggingface.co/prithivMLmods/Jan-v1-AIO-GGUF/tree/main/Jan-v1-edge-GGUF) |
|
| 23 |
+
| **Jan-v1-4B-GGUF** | [🔗 Link](https://huggingface.co/prithivMLmods/Jan-v1-AIO-GGUF/tree/main/Jan-v1-4B-GGUF) |
|
| 24 |
+
|
| 25 |
+
## Model Files
|
| 26 |
+
|
| 27 |
+
### Jan-v1-edge
|
| 28 |
+
|
| 29 |
+
| File Name | Quant Type | File Size |
|
| 30 |
+
| - | - | - |
|
| 31 |
+
| Jan-v1-edge.BF16.gguf | BF16 | 3.45 GB |
|
| 32 |
+
| Jan-v1-edge.F16.gguf | F16 | 3.45 GB |
|
| 33 |
+
| Jan-v1-edge.F32.gguf | F32 | 6.89 GB |
|
| 34 |
+
| Jan-v1-edge.Q2_K.gguf | Q2_K | 778 MB |
|
| 35 |
+
| Jan-v1-edge.Q3_K_L.gguf | Q3_K_L | 1 GB |
|
| 36 |
+
| Jan-v1-edge.Q3_K_M.gguf | Q3_K_M | 940 MB |
|
| 37 |
+
| Jan-v1-edge.Q3_K_S.gguf | Q3_K_S | 867 MB |
|
| 38 |
+
| Jan-v1-edge.Q4_0.gguf | Q4_0 | 1.05 GB |
|
| 39 |
+
| Jan-v1-edge.Q4_1.gguf | Q4_1 | 1.14 GB |
|
| 40 |
+
| Jan-v1-edge.Q4_K.gguf | Q4_K | 1.11 GB |
|
| 41 |
+
| Jan-v1-edge.Q4_K_M.gguf | Q4_K_M | 1.11 GB |
|
| 42 |
+
| Jan-v1-edge.Q4_K_S.gguf | Q4_K_S | 1.06 GB |
|
| 43 |
+
| Jan-v1-edge.Q5_0.gguf | Q5_0 | 1.23 GB |
|
| 44 |
+
| Jan-v1-edge.Q5_1.gguf | Q5_1 | 1.32 GB |
|
| 45 |
+
| Jan-v1-edge.Q5_K.gguf | Q5_K | 1.26 GB |
|
| 46 |
+
| Jan-v1-edge.Q5_K_M.gguf | Q5_K_M | 1.26 GB |
|
| 47 |
+
| Jan-v1-edge.Q5_K_S.gguf | Q5_K_S | 1.23 GB |
|
| 48 |
+
| Jan-v1-edge.Q6_K.gguf | Q6_K | 1.42 GB |
|
| 49 |
+
| Jan-v1-edge.Q8_0.gguf | Q8_0 | 1.83 GB |
|
| 50 |
+
|
| 51 |
+
### Jan-v1-4B
|
| 52 |
+
|
| 53 |
+
| File Name | Quant Type | File Size |
|
| 54 |
+
| - | - | - |
|
| 55 |
+
| Jan-v1-4B.BF16.gguf | BF16 | 8.05 GB |
|
| 56 |
+
| Jan-v1-4B.F16.gguf | F16 | 8.05 GB |
|
| 57 |
+
| Jan-v1-4B.F32.gguf | F32 | 16.1 GB |
|
| 58 |
+
| Jan-v1-4B.Q2_K.gguf | Q2_K | 1.67 GB |
|
| 59 |
+
| Jan-v1-4B.Q3_K_L.gguf | Q3_K_L | 2.24 GB |
|
| 60 |
+
| Jan-v1-4B.Q3_K_M.gguf | Q3_K_M | 2.08 GB |
|
| 61 |
+
| Jan-v1-4B.Q3_K_S.gguf | Q3_K_S | 1.89 GB |
|
| 62 |
+
| Jan-v1-4B.Q4_K_M.gguf | Q4_K_M | 2.5 GB |
|
| 63 |
+
| Jan-v1-4B.Q4_K_S.gguf | Q4_K_S | 2.38 GB |
|
| 64 |
+
| Jan-v1-4B.Q5_K_M.gguf | Q5_K_M | 2.89 GB |
|
| 65 |
+
| Jan-v1-4B.Q5_K_S.gguf | Q5_K_S | 2.82 GB |
|
| 66 |
+
| Jan-v1-4B.Q6_K.gguf | Q6_K | 3.31 GB |
|
| 67 |
+
| Jan-v1-4B.Q8_0.gguf | Q8_0 | 4.28 GB |
|
| 68 |
+
|
| 69 |
+
## Quants Usage
|
| 70 |
+
|
| 71 |
+
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
|
| 72 |
+
|
| 73 |
+
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
| 74 |
+
types (lower is better):
|
| 75 |
+
|
| 76 |
+
|