Hypernova 60B 2605 — Hardware Requirements & GPU Compatibility
Chat
Hypernova 60B 2605 is a 58.7B-parameter open language model from MultiverseComputingCAI. It supports a context window of up to 131,072 tokens. At Q4_K_M it needs about 35.59 GB of VRAM — see which GPUs and Macs can run it below.
Doesn't fit a 24 GB GPU
Smallest consumer fit → Snapdragon X Elite Copilot+ PC · Q2_K · 25.6 GB · ~37 tok/s
Macs from 32 GB (Q2_K · ~72 tok/s)
Apache 2.0
Not set up for it?
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Hypernova 60B 2605
One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.
Recommended
vllm serve MultiverseComputingCAI/Hypernova-60B-2605 --max-model-len 16384vLLM repo with its own quantisation
Download: Weights on Hugging Face
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on A40 (48 GB) (12 GB headroom) · Q4_K_M
- Generation speed
- ~121 tok/s
- generation speed
- Cost per 1M output tokens
- $0.81
- per 1M output tokens
How Much VRAM Does Hypernova 60B 2605 Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 25.3 GB | 26 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Does not fit: needs 25.6 GB | |
| 29.0 GB | 29.7 GB | 3-bit medium — 3-bit medium quantization | Does not fit: needs 29.3 GB | |
| 35.6 GB | 36.3 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Does not fit: needs 35.9 GB | |
| 42.2 GB | 42.9 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Does not fit: needs 42.5 GB | |
| 48.8 GB | 49.5 GB | 6-bit, very good — 6-bit quantization, very good quality | Does not fit: needs 49.1 GB | |
| 59.0 GB | 59.7 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Does not fit: needs 59.3 GB | |
| 117.7 GB | 118.4 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Does not fit: needs 118 GB |
Selected Q4_K_M
4-bit medium quantization35.9GBat 16K context
- Weights
- 35.2 GB
- KV cache
- 0.4 GB
- Overhead
- 0.3 GB
Doesn't fit RTX 4090 (reference) · needs 11.9 GB more
Or .
Which Hardware Can Run Hypernova 60B 2605?
Q4_K_M · 35.6 GBHypernova 60B 2605 (Q4_K_M) requires 35.6 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 47+ GB is recommended. Using the full 131K context window can add up to 2.9 GB, bringing total usage to 38.5 GB. No consumer GPU has enough memory. 67 devices with unified memory can run Hypernova 60B 2605, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Studio M5 Max (48 GB).
32–40 GB
Q4_K_M · best fit Q3_K_M (29.3 GB)
1 GPU · 2 Mac classes · 1 device family
RTX 5090~261Simulate output on RTX 5090M5 Max36 GB~72Simulate output on Mac Studio M5 Max (36 GB)M3 Pro*36 GB~34Simulate output on MacBook Pro 14" M3 Pro (36 GB)Orin~50Simulate output on NVIDIA Jetson AGX Orin 32GB48 GB
48 GB tier →fits · 12.1 GB free · up to 128K ctx
2 Mac classes · 1 device family
M5 Max48 GB~75Simulate output on Mac Studio M5 Max (48 GB)M5 Pro48 GB~50Simulate output on Mac Mini M5 Pro (48 GB)Snapdragon X2 Elite Extreme (X2E-96-100)~43Simulate output on Snapdragon X2 Elite Extreme Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Similar models
2| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Qwen2 57B A14B InstructAlibaba | — | 35.7 GB | —needs > 24 GB | same size |
| Mixtral 34Bx2 MoE 60Bcloudyu | — | 40.8 GB | —needs > 24 GB | same size |
Specifications
- Publisher
- MultiverseComputingCAI
- Parameters
- 58.7B
- Active parameters
- 4.9B per token
- Architecture
- GptOssForCausalLM
- Context Length
- 131,072 tokens
- Vocabulary Size
- 201,088
- Release Date
- 2026-04-30
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does Hypernova 60B 2605 need?
Hypernova 60B 2605 requires 35.6 GB of VRAM at Q4_K_M, or 117.7 GB at BF16. Full 131K context adds up to 2.9 GB (38.5 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 58.7B × 4.8 bits ÷ 8 = 35.2 GB
KV Cache + Overhead ≈ 0.4 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 ≈ 3.3 GB (at full 131K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 5090 run Hypernova 60B 2605?
Yes, at Q3_K_M (29.0 GB) or lower. Higher quantizations like Q4_K_M (35.6 GB) exceed the NVIDIA GeForce RTX 5090's 32 GB.
What's the best quantization for Hypernova 60B 2605?
For Hypernova 60B 2605, Q4_K_M (35.6 GB) offers the best balance of quality and VRAM usage. Q5_K_M (42.2 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 25.3 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Hypernova 60B 2605 on a Mac?
Yes — Mac Studio M1 Max (32 GB) and 84 other Macs can run Hypernova 60B 2605, the smallest at Q2_K. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 35.6 GB of usable unified memory (RAM minus macOS overhead), from Mac Studio M5 Max (48 GB).
Can I run Hypernova 60B 2605 locally?
Yes — Hypernova 60B 2605 can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Hypernova 60B 2605?
At Q4_K_M, Hypernova 60B 2605 can reach ~143 tok/s on AMD Instinct MI350X. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = 1000 ÷ (active GB ÷ (bandwidth GB/s × efficiency) × 1000 + layers × routing ms)
Mixture-of-Experts: just the active experts get read for each token, along with a fixed routing cost per layer.
Example: NVIDIA B200 → 3.0 GB active ÷ (8000 × 0.65) = 0.568 ms, plus 32 layers × 0.055 ms = 1.760 ms, so 1000 ÷ 2.328 ms = ~430 tok/s
Estimated speed at Q4_K_M (35.6 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Hypernova 60B 2605?
At Q4_K_M, the download is about 35.20 GB. The full-precision BF16 version is 117.32 GB. The smallest option (Q2_K) is 24.93 GB.
Which GPUs can run Hypernova 60B 2605?
No single consumer GPU has enough VRAM to run Hypernova 60B 2605 at Q4_K_M (35.6 GB). Multi-GPU or professional hardware is required.
Which devices can run Hypernova 60B 2605?
67 devices with unified memory can run Hypernova 60B 2605 at Q4_K_M (35.6 GB), including ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB), GMKtec EVO-X2 (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.