# Darwin-180B-RSI-R3

> 💻 Run it on your own machine: POCKET-Darwin-180B-GGUF, the 4-bit GGUF of R3 (111 GB), runs on a laptop with an 8 GB GPU and 32 GB RAM, CPU only at 18–21 tok/s, a 128 GB mini PC or one DGX Spark. MMLU-Pro is identical to BF16 (87.65%).

- **Organisation:** FINAL-Bench
- **Category:** Vision-Language Models, Reasoning Models
- **Parameters:** 180B total, 61.6B active (MoE)
- **Context length:** 262,144 tokens
- **Licence:** qwen-community-1.0 (Commercial use with conditions)
- **Released:** 2026-10-01
- **Hugging Face:** 1.1K downloads (30 days), 49 likes

## Estimated memory

| Quant | ≈ GB |
|---|---|
| Q2_K | 59.12 |
| Q3_K_M | 87.28 |
| Q4_K_M | 107.96 |
| Q5_K_M | 126.44 |
| Q6_K | 146.25 |
| Q8_0 | 188.27 |
| F16 | 353.29 |

## Runs with

- [llama.cpp](https://selfhost.directory/project/llamacpp)
- [LM Studio](https://lmstudio.ai)
- [SGLang](https://selfhost.directory/project/sglang)
- [Transformers](https://github.com/huggingface/transformers)
- [vLLM](https://selfhost.directory/project/vllm)
- [ExLlamaV2](https://github.com/turboderp-org/exllamav2) (community)
- [Jan](https://jan.ai) (community)
- [KoboldCpp](https://selfhost.directory/project/koboldcpp) (community)
- [LocalAI](https://selfhost.directory/project/localai) (community)
- [MLX](https://github.com/ml-explore/mlx-lm) (community)
- [Ollama](https://selfhost.directory/project/ollama) (community)

## About

💻 Run it on your own machine: POCKET-Darwin-180B-GGUF, the 4-bit GGUF of R3 (111 GB), runs on a laptop with an 8 GB GPU and 32 GB RAM, CPU only at 18–21 tok/s, a 128 GB mini PC or one DGX Spark. MMLU-Pro is identical to BF16 (87.65%).

reasoning · MoE 512 experts · 262K long context · image + text · Korean + English · self-improvement · structured output · ZTC

The second round of model-level self-improvement on top of Darwin-180B-RSI. R3 continues training from R1 with the same recipe: the model solves verifiable problems, keeps only its own solutions that check out as correct, and trains on them. No human-written solutions or reasoning traces are used. R3 is released so anyone can download it, run it, and check the numbers below.

---

Hugging Face official benchmark leaderboards. Each score is listed under the model that produced it.

R3's own leaderboard entries:

Full settings are recorded in .eval_results/.

Scores as listed on the Hugging Face official benchmark leaderboards (self-reported by each model's publisher). "·" = not reported. Open-weight models only; closed API models are not included. Settings (samples, voting, thinking budget) differ across models. R1 settings are in the evaluation protocol below.

---

## Alternatives

- [Qwen/Qwen3.8-Flash-Next](https://selfhost.directory/models/qwen/qwen3.8-flash-next) (180B (61.6B active), qwen-community-1.0, ≈107.96 GB at Q4): As the frontier of foundation models pushes toward ever-larger parameter counts and ever-longer context windows, the question is no longer just how much we can scale, but how efficiently we can do so.
- [MiniMaxAI/MiniMax-M3.1-Flash](https://selfhost.directory/models/minimaxai/minimax-m3.1-flash) (220B, minimax-community, ≈132.15 GB at Q4): MiniMax-M3.1-Flash is a native multimodal mixture-of-experts model for coding, agentic workflows, computer use, and cybersecurity. It supports text, image, and video inputs with a context window of up to 1M tokens.
- [zai-org/GLM-5.3-Flash](https://selfhost.directory/models/zai-org/glm-5.3-flash) (321B (18.3B active), MIT, ≈213.49 GB at Q4): 👋 Join our WeChat or Discord community. 📖 Check out the GLM-5.3-Flash blog and GLM-5 Technical report . 📍 Use GLM-5.3-Flash API services on Z.ai API Platform.
- [thinkingmachines/Inkling-Small](https://selfhost.directory/models/thinkingmachines/inkling-small) (266B (1.7B active), Apache 2.0, ≈159.48 GB at Q4): Inkling-Small is a general-purpose multimodal model that accepts text, image and audio inputs and generates text outputs. It is intended for use in English and other languages, and across multiple coding languages.
- [Qwen/Qwen3.5-122B-A10B](https://selfhost.directory/models/qwen/qwen3.5-122b-a10b) (125B (10B active), Apache 2.0, ≈75.41 GB at Q4): Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
- [nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16](https://selfhost.directory/models/nvidia/nvidia-nemotron-3.5-super-vl-120b-a12b-bf16) (124B (12B active), openmdw-1.1, ≈75.54 GB at Q4): NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 is the full-precision release of NVIDIA Nemotron 3.5 Super VL, a vision-language model for reasoning, coding, tool use, document understanding, and video understanding.

More: https://selfhost.directory/models/final-bench/darwin-180b-rsi-r3/alternatives

## Runs well on

- [Apple M2 Ultra (192 GB)](https://selfhost.directory/models/final-bench/darwin-180b-rsi-r3/on/apple-m2-ultra-192gb): Q5_K_M, est. 10 tok/s
- [Apple M3 Ultra (512 GB)](https://selfhost.directory/models/final-bench/darwin-180b-rsi-r3/on/apple-m3-ultra-512gb): F16, est. 3.7 tok/s

## Similar size

- [Qwen/Qwen3.8-Flash-Next](https://selfhost.directory/models/qwen/qwen3.8-flash-next) (180B)
- [MiniMaxAI/MiniMax-M3.1-Flash](https://selfhost.directory/models/minimaxai/minimax-m3.1-flash) (220B)
- [tencent/Hy4-flash](https://selfhost.directory/models/tencent/hy4-flash) (220B)
- [stepfun-ai/Step-3.7-Flash](https://selfhost.directory/models/stepfun-ai/step-3.7-flash) (201B)
- [CohereLabs/command-a-plus-05-2026-bf16](https://selfhost.directory/models/coherelabs/command-a-plus-05-2026-bf16) (219B)
- [FINAL-Bench/Darwin-180B-RSI](https://selfhost.directory/models/final-bench/darwin-180b-rsi) (180B)

## FAQ

**How much VRAM does Darwin-180B-RSI-R3 need?**
About 108 GB at Q4_K_M with an 8K context. Q8_0 needs about 188 GB and full precision about 353 GB. Longer contexts need more memory for the KV cache.

**Can Darwin-180B-RSI-R3 run on the Apple M2 Ultra (192 GB)?**
Yes. It fits at Q5_K_M (about 126 GB), at an estimated 10 tokens/s. It fits comfortably on the Apple M3 Ultra (512 GB).

**Can I use Darwin-180B-RSI-R3 commercially?**
Yes, with conditions. It is released under qwen-community-1.0. Custom licence: read the model card before using it commercially.

**What is the context length of Darwin-180B-RSI-R3?**
262,144 tokens (256K). The full window needs far more memory than an 8K context.

**How do I run Darwin-180B-RSI-R3 with Ollama or llama.cpp?**
With Ollama: ollama run hf.co/FINAL-Bench/POCKET-Darwin-180B-GGUF:UD-Q4_K_XL. With llama.cpp: llama-server -hf FINAL-Bench/POCKET-Darwin-180B-GGUF:UD-Q4_K_XL, which starts an OpenAI-compatible server.

## Links

- huggingface: https://huggingface.co/FINAL-Bench/Darwin-180B-RSI-R3
- paper: https://arxiv.org/abs/2609.20269

---
Source: https://selfhost.directory/models/final-bench/darwin-180b-rsi-r3