Kimi K2 Instruct 0905 — Hardware Requirements & GPU Compatibility
Chat
Kimi-K2-Instruct-0905 is Moonshot AI's updated flagship chat and agentic-coding model, a mixture-of-experts system with roughly 1.03 trillion total parameters and about 32.9 billion active per token, routed across 384 experts with 8 selected plus one shared expert per token. It is tuned for agentic coding, tool calling, and long-horizon agent workflows, using multi-head latent attention and SwiGLU activations, and Moonshot reports clear gains on coding-agent benchmarks like SWE-bench and Terminal-Bench, plus improved frontend code aesthetics, over the prior K2-Instruct-0711 release. With a trillion-parameter total footprint it needs a multi-GPU server-class setup even once quantized. Context length is 262,144 tokens, extended from 128K in the previous release. It is released under Moonshot's Modified MIT License, which is otherwise permissive but requires products with more than 100 million monthly active users or $20 million in monthly revenue to display "Kimi K2" in their interface. It was published in September 2025.
55.4K downloads
788 likesSpecifications
- Publisher
- Moonshot AI
- Family
- Kimi K2
- Parameters
- 1026.5B
- Architecture
- DeepseekV3ForCausalLM
- Context Length
- 262,144 tokens
- Vocabulary Size
- 163,840
- Release Date
- 2025-09-03
- License
- Other
Get Started
HuggingFace
How Much VRAM Does Kimi K2 Instruct 0905 Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_K | 3.40 | 440.1 GB | 895.0 GB | 436.25 GB | 2-bit quantization with K-quant improvements |
| Q3_K_S | 3.50 | 453.0 GB | 907.9 GB | 449.08 GB | 3-bit small quantization |
| Q3_K_M | 3.90 | 504.3 GB | 959.2 GB | 500.40 GB | 3-bit medium quantization |
| Q4_0 | 4.00 | 517.1 GB | 972.0 GB | 513.24 GB | 4-bit legacy quantization |
| Q4_K_M | 4.80 | 619.8 GB | 1074.7 GB | 615.88 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_M | 5.70 | 735.2 GB | 1190.2 GB | 731.36 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 850.7 GB | 1305.6 GB | 846.84 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 1030.3 GB | 1485.3 GB | 1026.47 GB | 8-bit quantization, near-lossless |
Which GPUs Can Run Kimi K2 Instruct 0905?
Q4_K_M · 619.8 GBKimi K2 Instruct 0905 (Q4_K_M) requires 619.8 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 806+ GB is recommended. Using the full 262K context window can add up to 454.9 GB, bringing total usage to 1074.7 GB. No single GPU has enough memory — multi-GPU or cluster setups are needed.
Which Devices Can Run Kimi K2 Instruct 0905?
Q4_K_M · 619.8 GB2 devices with unified memory can run Kimi K2 Instruct 0905, including NVIDIA DGX H100.
Decent
— Enough memory, may be tightWhere to Download Kimi K2 Instruct 0905
Community quantizations of this model — GGUF for llama.cpp, Ollama, and LM Studio, plus AWQ/MLX variants where available.
Benchmarks
Benchmark details →Related Models
Frequently Asked Questions
- How much VRAM does Kimi K2 Instruct 0905 need?
Kimi K2 Instruct 0905 requires 619.8 GB of VRAM at Q4_K_M, or 2056.8 GB at BF16. Full 262K context adds up to 454.9 GB (1074.7 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 1026.5B × 4.8 bits ÷ 8 = 615.9 GB
KV Cache + Overhead ≈ 3.9 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 458.8 GB (at full 262K context)
VRAM usage by quantization
Q4_K_M619.8 GBQ4_K_M + full context1074.7 GB- Can NVIDIA GeForce RTX 5090 run Kimi K2 Instruct 0905?
No — Kimi K2 Instruct 0905 requires at least 286.2 GB at IQ2_XXS, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
- What's the best quantization for Kimi K2 Instruct 0905?
For Kimi K2 Instruct 0905, Q4_K_M (619.8 GB) offers the best balance of quality and VRAM usage. Q5_K_S (709.6 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 286.2 GB.
VRAM requirement by quantization
IQ2_XXS286.2 GBIQ3_XS427.3 GBQ3_K_L530.0 GBQ4_K_M ★619.8 GBQ5_K_S709.6 GBBF162056.8 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run Kimi K2 Instruct 0905 on a Mac?
Kimi K2 Instruct 0905 requires at least 286.2 GB at IQ2_XXS, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run Kimi K2 Instruct 0905 locally?
Yes — Kimi K2 Instruct 0905 can run locally on consumer hardware. At Q4_K_M quantization it needs 619.8 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- What's the download size of Kimi K2 Instruct 0905?
At Q4_K_M, the download is about 615.88 GB. The full-precision BF16 version is 2052.94 GB. The smallest option (IQ2_XXS) is 282.28 GB.
- Which GPUs can run Kimi K2 Instruct 0905?
No single consumer GPU has enough VRAM to run Kimi K2 Instruct 0905 at Q4_K_M (619.8 GB). Multi-GPU or professional hardware is required.
- Which devices can run Kimi K2 Instruct 0905?
2 devices with unified memory can run Kimi K2 Instruct 0905 at Q4_K_M (619.8 GB), including NVIDIA DGX A100 640GB, NVIDIA DGX H100. Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.