OneReason 0.8B Pretrain Competition — Hardware Requirements & GPU Compatibility
ChatReasoning
OneReason 0.8B Pretrain Competition is a 801M-parameter open language model from OpenOneRec. It supports a context window of up to 40,960 tokens. At Q4_K_M it needs about 0.90 GB of VRAM — see which GPUs and Macs can run it below.
33.5K downloads
26 likesSpecifications
- Publisher
- OpenOneRec
- Parameters
- 801M
- Architecture
- Qwen3ForCausalLM
- Context Length
- 40,960 tokens
- Vocabulary Size
- 176,253
- Release Date
- 2026-06-09
Get Started
How Much VRAM Does OneReason 0.8B Pretrain Competition Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_Kest. | 3.40 | 0.8 GB | 3.0 GB | 0.34 GB | 2-bit quantization with K-quant improvements |
| Q3_K_Mest. | 3.90 | 0.8 GB | 3.0 GB | 0.39 GB | 3-bit medium quantization |
| Q4_K_Mest. | 4.80 | 0.9 GB | 3.1 GB | 0.48 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_Mest. | 5.70 | 1.0 GB | 3.2 GB | 0.57 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_Kest. | 6.60 | 1.1 GB | 3.3 GB | 0.66 GB | 6-bit quantization, very good quality |
| Q8_0est. | 8.00 | 1.2 GB | 3.5 GB | 0.80 GB | 8-bit quantization, near-lossless |
| BF16est. | 16.00 | 2.0 GB | 4.3 GB | 1.60 GB | Brain floating point 16 — preferred for training |
est.= calculated VRAM estimate; no published GGUF file found for that quantization yet. Other rows are verified against real community uploads.
Which GPUs Can Run OneReason 0.8B Pretrain Competition?
Q4_K_M · 0.9 GBOneReason 0.8B Pretrain Competition (Q4_K_M) requires 0.9 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 2+ GB is recommended. Using the full 41K context window can add up to 2.2 GB, bringing total usage to 3.1 GB. 50 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.
Runs great
— Plenty of headroomWhich Devices Can Run OneReason 0.8B Pretrain Competition?
Q4_K_M · 0.9 GB59 devices with unified memory can run OneReason 0.8B Pretrain Competition, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
Runs great
— Plenty of headroomFrequently Asked Questions
- How much VRAM does OneReason 0.8B Pretrain Competition need?
OneReason 0.8B Pretrain Competition requires 0.9 GB of VRAM at Q4_K_M, or 2.0 GB at BF16. Full 41K context adds up to 2.2 GB (3.1 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 801M × 4.8 bits ÷ 8 = 0.5 GB
KV Cache + Overhead ≈ 0.4 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 2.6 GB (at full 41K context)
VRAM usage by quantization
Q4_K_M0.9 GBQ4_K_M + full context3.1 GB- What's the best quantization for OneReason 0.8B Pretrain Competition?
For OneReason 0.8B Pretrain Competition, Q4_K_M (0.9 GB) offers the best balance of quality and VRAM usage. Q5_K_M (1.0 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 0.8 GB.
VRAM requirement by quantization
Q2_K0.8 GBQ4_K_M ★0.9 GBQ5_K_M1.0 GBQ6_K1.1 GBQ8_01.2 GBBF162.0 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run OneReason 0.8B Pretrain Competition on a Mac?
OneReason 0.8B Pretrain Competition requires at least 0.8 GB at Q2_K, 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 OneReason 0.8B Pretrain Competition locally?
Yes — OneReason 0.8B Pretrain Competition can run locally on consumer hardware. At Q4_K_M quantization it needs 0.9 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is OneReason 0.8B Pretrain Competition?
At Q4_K_M, OneReason 0.8B Pretrain Competition can reach ~4889 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~728 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 ÷ 0.9 × 0.65 = ~5778 tok/s
Estimated speed at Q4_K_M (0.9 GB)
~5778 tok/s~728 tok/s~5778 tok/s~4889 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of OneReason 0.8B Pretrain Competition?
At Q4_K_M, the download is about 0.48 GB. The full-precision BF16 version is 1.60 GB. The smallest option (Q2_K) is 0.34 GB.
- Which GPUs can run OneReason 0.8B Pretrain Competition?
50 consumer GPUs can run OneReason 0.8B Pretrain Competition at Q4_K_M (0.9 GB). Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 50 GPUs have plenty of headroom for comfortable inference.
- Which devices can run OneReason 0.8B Pretrain Competition?
59 devices with unified memory can run OneReason 0.8B Pretrain Competition at Q4_K_M (0.9 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.