NVIDIA Nemotron™ is a family of open models with open weights, training data, and recipes, delivering leading efficiency and accuracy for building specialized AI agents.
Darwin-27B-RSI
FINAL-Bench·Instruct·released 27 Sept 2026
ollama run hf.co/FINAL-Bench/Darwin-27B-RSI-GGUF:Q4_K_MQwen3.5-27B family · 27B dense · Thinking mode · BF16 · Apache 2.0 No human-written answers. The model generated its own learning signal — and got measurably better.
How much VRAM does Darwin-27B-RSI need?
At Q4_K_M Darwin-27B-RSI needs about 18 GB with an 8K context (18 GB at Q4, 55 GB at full precision). Pick your GPU to see what fits and how fast it runs.
Needs: ≈19 GB · Q4_K_M · 8K context — Pick a GPU to check the fit.
Advanced: context 8K · KV cache 16-bit
Smallest GPU for each quant and context (KV cache 16-bit)
| Quantisation | 8K | 16K | 32K | 64K | 128K | 256K |
|---|---|---|---|---|---|---|
| Q3_K_M | ||||||
| Q4_K_M | ||||||
| Q5_K_M | ||||||
| Q6_K | ||||||
| Q8_0 | ||||||
| FP16 |
≤16 GB card24–32 GB48–96 GBmore
| Quantisation | Total |
|---|---|
| Q2_KHeavy lossSmallest, noticeably worse quality | ≈13 GB |
| Q3_K_MNoticeable lossSmall, some quality loss | ≈15 GB |
| Q4_K_MrecommendedSmall lossRecommended default for local use | ≈19 GB |
| Q5_K_MNear-losslessNear-lossless for most tasks | ≈23 GB |
| Q6_KNear-losslessPractically lossless | ≈26 GB |
| Q8_0LosslessLossless in practice, twice the size of Q4 | ≈31 GB |
| FP16 / BF16LosslessFull precision, as released | ≈55 GB |
Estimates for 8K tokens of context: weights + KV cache + runtime overhead. Speed assumes memory-bound decoding at your GPU’s bandwidth. Pick a GPU above and the fit column checks every row against it. Weights marked “file” use the real GGUF download size; “est.” rows are computed from the parameter count.
How to pick a quant
Take the biggest quant that fits with room left for context. Q4_K_M is the usual sweet spot. Below Q4 quality drops fast, above Q6 you mostly pay memory for nothing. Big models (70B+, large MoE) tolerate low quants better than small ones.
- LosslessF16, Q8_0
- Near-losslessQ6_K, Q5_K_M
- Small lossQ5_K_S, Q4_K_M, IQ4_NL, IQ4_XS, Q4_K_S
- Noticeable lossQ4_0, Q3_K_L, Q3_K_M, IQ3_M
- Heavy lossIQ3_XXS, Q3_K_S, IQ2_M, Q2_K
Typical quality loss, not measured per model. Tiers come from KL divergence against FP16 on Llama 3 8B: llama.cpp perplexity README: LLaMA 3 8B scoreboard (KLD and ΔPPL vs FP16, Wikitext-2); llama.cpp quantize tool: ΔPPL per quant type @ Llama-3-8B
How much context fits?
At 8K context, Q4_K_M needs ≈19 GB(16 GB weights + 2.0 GB KV cache + 1.3 GB overhead). Pick a GPU above to see how much context fits.
Weights + overheadKV cache
Can the Tesla P40 run Darwin-27B-RSI? →
Speed vs quality per quant
Smaller quants run faster and fit in less memory, but lose quality.
Fits in VRAMTight fitCPU offload✕ won't runQuality is typical KL divergence vs FP16, not measured for this model.
Real-world speed of Darwin-27B-RSI
Tokens per second measured by people running it at home. A GPU shows a median once 3 reports agree on the same setup. Reports far from our estimate wait for a review. Measure yours with llama-bench →
No medians yet. Be one of the first 3 to report a setup.
All reports
Run Darwin-27B-RSI locally
--quantQ4_K_Mdefault; pick your GPU to tailor itFINAL-Bench/Darwin-27B-RSI-GGUF
ollama run hf.co/FINAL-Bench/Darwin-27B-RSI-GGUF:Q4_K_M llama-server -hf FINAL-Bench/Darwin-27B-RSI-GGUF:Q4_K_MDownloads the GGUF on first run. Installed through llama.app? The same flags work with `llama serve`.
lms get FINAL-Bench/Darwin-27B-RSI-GGUF@Q4_K_MThen load it from the LM Studio app or with `lms load`.
vllm serve FINAL-Bench/Darwin-27B-RSIservices:
vllm:
image: vllm/vllm-openai:latest
command: FINAL-Bench/Darwin-27B-RSI
ports:
- "8000:8000"
ipc: host
volumes:
- ~/.cache/huggingface:/root/.cache/huggingface
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]API on port 8000.
sglang serve FINAL-Bench/Darwin-27B-RSI --host 0.0.0.0 --port 30000services:
sglang:
image: lmsysorg/sglang:latest
command: sglang serve FINAL-Bench/Darwin-27B-RSI --host 0.0.0.0 --port 30000
ports:
- "30000:30000"
ipc: host
volumes:
- ~/.cache/huggingface:/root/.cache/huggingface
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]API on port 30000.
from transformers import pipeline
pipe = pipeline("text-generation", model="FINAL-Bench/Darwin-27B-RSI", dtype="auto", device_map="auto")
messages = [{"role": "user", "content": "Explain self-hosting in one paragraph."}]
out = pipe(messages, max_new_tokens=512)
print(out[0]["generated_text"][-1]["content"])Quantised versions and downloads
| Repository | Format | Quant | Size |
|---|---|---|---|
| bartowski/FINAL-Bench_Darwin-27B-RSI-GGUF | GGUF | IQ2_XXS | 8.6 GB |
| bartowski/FINAL-Bench_Darwin-27B-RSI-GGUF | GGUF | IQ2_XS | 8.8 GB |
| bartowski/FINAL-Bench_Darwin-27B-RSI-GGUF | GGUF | IQ2_S | 9.4 GB |
| bartowski/FINAL-Bench_Darwin-27B-RSI-GGUF | GGUF | IQ2_M | 10 GB |
| bartowski/FINAL-Bench_Darwin-27B-RSI-GGUF | GGUF | Q2_K | 11 GB |
| bartowski/FINAL-Bench_Darwin-27B-RSI-GGUF | GGUF | IQ3_XXS | 12 GB |
| bartowski/FINAL-Bench_Darwin-27B-RSI-GGUF | GGUF | Q3_K_S | 13 GB |
| bartowski/FINAL-Bench_Darwin-27B-RSI-GGUF | GGUF | IQ3_XS | 13 GB |
| bartowski/FINAL-Bench_Darwin-27B-RSI-GGUF | GGUF | Q3_K_M | 13 GB |
| bartowski/FINAL-Bench_Darwin-27B-RSI-GGUF | GGUF | Q3_K_L | 14 GB |
About Darwin-27B-RSI
Qwen3.5-27B family · 27B dense · Thinking mode · BF16 · Apache 2.0 No human-written answers. The model generated its own learning signal — and got measurably better. --- Darwin-27B-RSI is Darwin-27B-Opus after Recursive Self-Improvement (RSI): the model was improved using only signal it produced itself. During self-improvement, the model itself (its weights) improves by learning only from its own solutions. No human-written solutions or reasoning traces are used; correctness is checked automatically (agreement across its own samples and code execution). Under an identical evaluation protocol, Darwin-27B-RSI improves over its parent on graduate-level science reasoning — +5.24 points on GPQA Diamond (single sample) and +3.79 points with majority voting — with every gain statistically significant in paired tests. As the reasoning engine of Darwin-27B-JEV on the Decision Index, it lifts the hardest reasoning decisions: GPQA Diamond skill 0.31 → 0.71, GSM8K 0.61 → 0.97, MMLU-Pro 0.60 → 0.82. --- Darwin-27B-RSI is Model-level RSI: the model itself (its weights) improves by learning only from its own solutions. No human-written solutions or reasoning traces are used; correctness is checked automatically. You download a new model file, and it is smarter on its own.
Full model card on Hugging Face →Alternatives to Darwin-27B-RSI
All alternatives →Versioning notice: This card describes the current OTel 2.0 OSFT checkpoint. The model weights may be updated over time. For reproducible evaluation or production deployment, pin a specific model revision, checkpoint hash, or release tag rather than relying on the floating latest version.
Aloha! 🌺 Today, we are releasing Ornith-1.0, a self-improving family of open-source models for agentic coding.
👋 Join our Discord community. 📖 Check out the GLM-4.7 technical blog , technical report(GLM-4.5) . 📍 Use GLM-4.7-Flash API services on Z.ai API Platform. 👉 One click to GLM-4.7 .
Laguna XS 2.1 is a 33B total parameter Mixture-of-Experts model with 3B activated parameters per token designed for agentic coding and long-horizon work on a local machine.
Nemotron-3-Nano-30B-A3B-BF16 is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks. It responds to user queries and tasks by first generating a reasoning trace and then concluding with a final response.
Darwin-27B-RSI FAQ
- How much VRAM does Darwin-27B-RSI need?
- About 18 GB at Q4_K_M with an 8K context. Q8_0 needs about 30 GB and full precision about 55 GB. Longer contexts need more memory for the KV cache.
- Can Darwin-27B-RSI run on the Tesla P40?
- Yes. It fits at Q5_K_M (about 23 GB), at an estimated 10 tokens/s. It fits comfortably on the Radeon RX 7900 XTX, Instinct MI50 32GB, RTX 5090.
- Can I use Darwin-27B-RSI commercially?
- Yes. It is released under Apache 2.0. Permissive: use, modify and sell freely; keep the notice and licence text.
- What is the context length of Darwin-27B-RSI?
- 262,144 tokens (256K). The full window needs far more memory than an 8K context.
- How do I run Darwin-27B-RSI with Ollama or llama.cpp?
- With Ollama: ollama run hf.co/FINAL-Bench/Darwin-27B-RSI-GGUF:Q4_K_M. With llama.cpp: llama-server -hf FINAL-Bench/Darwin-27B-RSI-GGUF:Q4_K_M, which starts an OpenAI-compatible server.
Rate this model
Data from Hugging Face, the model card and public leaderboards. Memory and speed figures are estimates; real numbers depend on the runtime, context length and drivers. Always check the licence yourself before commercial use.
0 comments