FrogNano 4B 2609 — Hardware Requirements & GPU Compatibility
ChatFunctions
FrogNano is Microsoft's compact repository-level coding agent, derived from Qwen3.5-4B and holding about 4.7 billion parameters. It is post-trained only with reinforcement learning on roughly 1,500 synthetic software-engineering tasks, using a lightweight five-tool harness called Leaf, and the card states that it uses no solution trajectories or reasoning traces from stronger models. It keeps the dense 32-layer hybrid Gated DeltaNet and gated-attention design of its base. Its intended use is text-only: repository navigation, debugging, code editing, test execution and patch generation. At this size it fits on almost any modern GPU and on many laptops, especially when quantized. The model configuration allows 262,144 tokens, while the card describes an evaluated agent setup with about 131K tokens of combined context. The card's summary table lists the Apache 2.0 license inherited from Qwen3.5-4B, while the repository is tagged MIT; both permit unrestricted commercial and research use. Published in September 2026, it is a specialist derivative of Qwen3.5-4B rather than a general-purpose chat model.
1.1K downloads
139 likesSpecifications
- Publisher
- Microsoft
- Parameters
- 4.7B
- Architecture
- Qwen3_5ForConditionalGeneration
- Context Length
- 262,144 tokens
- Vocabulary Size
- 248,320
- Release Date
- 2026-09-17
- License
- MIT
Get Started
HuggingFace
Run in cloud
Fits on RTX 3060 12GB (8 GB headroom) · Q4_K_M
- Generation speed
- ~72 tok/s
- generation speed
- Cost per 1M output tokens
- $0.23
- per 1M output tokens
How Much VRAM Does FrogNano 4B 2609 Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_K | 3.40 | 2.5 GB | 7.7 GB | 1.98 GB | 2-bit quantization with K-quant improvements |
| Q3_K_S | 3.50 | 2.5 GB | 7.7 GB | 2.04 GB | 3-bit small quantization |
| Q3_K_M | 3.90 | 2.7 GB | 7.9 GB | 2.27 GB | 3-bit medium quantization |
| Q4_0 | 4.00 | 2.8 GB | 8 GB | 2.33 GB | 4-bit legacy quantization |
| Q4_K_M | 4.80 | 3.3 GB | 8.5 GB | 2.80 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_M | 5.70 | 3.8 GB | 9.0 GB | 3.32 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 4.3 GB | 9.5 GB | 3.84 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 5.1 GB | 10.3 GB | 4.66 GB | 8-bit quantization, near-lossless |
Which GPUs Can Run FrogNano 4B 2609?
Q4_K_M · 3.3 GBFrogNano 4B 2609 (Q4_K_M) requires 3.3 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 5+ GB is recommended. Using the full 262K context window can add up to 5.2 GB, bringing total usage to 8.5 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.
Runs great
— Plenty of headroomWhich Devices Can Run FrogNano 4B 2609?
Q4_K_M · 3.3 GB59 devices with unified memory can run FrogNano 4B 2609, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, MacBook Air 13" M3 (8 GB).
Runs great
— Plenty of headroom
Decent
— Enough memory, may be tightWhere to Download FrogNano 4B 2609
Community quantizations of this model — GGUF for llama.cpp, Ollama, and LM Studio, plus AWQ/MLX variants where available.
Related Models
Frequently Asked Questions
- How much VRAM does FrogNano 4B 2609 need?
FrogNano 4B 2609 requires 3.3 GB of VRAM at Q4_K_M, or 9.8 GB at BF16. Full 262K context adds up to 5.2 GB (8.5 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 4.7B × 4.8 bits ÷ 8 = 2.8 GB
KV Cache + Overhead ≈ 0.5 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 5.7 GB (at full 262K context)
VRAM usage by quantization
Q4_K_M3.3 GBQ4_K_M + full context8.5 GB- What's the best quantization for FrogNano 4B 2609?
For FrogNano 4B 2609, Q4_K_M (3.3 GB) offers the best balance of quality and VRAM usage. Q4_K_L (3.3 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 1.8 GB.
VRAM requirement by quantization
IQ2_XXS1.8 GBIQ3_S2.5 GBQ3_K_L2.9 GBQ4_K_M ★3.3 GBQ5_03.4 GBBF169.8 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run FrogNano 4B 2609 on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 38 other Macs can run FrogNano 4B 2609. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 3.3 GB of usable unified memory (RAM minus macOS overhead).
- Can I run FrogNano 4B 2609 locally?
Yes — FrogNano 4B 2609 can run locally on consumer hardware. At Q4_K_M quantization it needs 3.3 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is FrogNano 4B 2609?
At Q4_K_M, FrogNano 4B 2609 can reach ~1472 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~201 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 3.26 × 0.65 = ~1595 tok/s
Estimated speed at Q4_K_M (3.3 GB)
~1595 tok/s~201 tok/s~1595 tok/s~1472 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of FrogNano 4B 2609?
At Q4_K_M, the download is about 2.80 GB. The full-precision BF16 version is 9.32 GB. The smallest option (IQ2_XXS) is 1.28 GB.
- Which GPUs can run FrogNano 4B 2609?
52 consumer GPUs can run FrogNano 4B 2609 at Q4_K_M (3.3 GB). Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 52 GPUs have plenty of headroom for comfortable inference.
- Which devices can run FrogNano 4B 2609?
59 devices with unified memory can run FrogNano 4B 2609 at Q4_K_M (3.3 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Apple iPhone 17 Pro, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.