Tev1 0.8B Experimental — Hardware Requirements & GPU Compatibility
ChatVisionTev1 0.8B Experimental is a 873M-parameter open language model from togethercomputer. It supports a context window of up to 262,144 tokens. At BF16 it needs about 2.10 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (BF16 · 256K)Best on 24 GB (BF16 · ~312 tok/s · 256K)Macs from 8 GB (BF16 · ~34 tok/s)
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Tev1 0.8B Experimental
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 togethercomputer/Tev1-0.8B-experimental --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 RTX 3060 12GB (9 GB headroom) · BF16
- Generation speed
- ~111 tok/s
- generation speed
- Cost per 1M output tokens
- $0.16
- per 1M output tokens
How Much VRAM Does Tev1 0.8B Experimental Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 2.1 GB | 2.3 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 256K context · ~312 tok/s |
Selected BF16
Brain floating point 162.2GBat 16K context
- Weights
- 1.8 GB
- KV cache
- 0.1 GB
- Overhead
- 0.3 GB
Which Hardware Can Run Tev1 0.8B Experimental?
BF16 · 2.1 GBTev1 0.8B Experimental (BF16) requires 2.1 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 3+ GB is recommended. Using the full 262K context window can add up to 1.6 GB, bringing total usage to 3.7 GB. 114 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 185 devices with unified memory can run Tev1 0.8B Experimental, including Mac Studio M5 Ultra (256 GB), Mac Studio M5 Ultra (512 GB).
≤ 11 GB
≤ 11 GB tier →fits · 1.9 GB free · up to 256K ctx
68 GPUs · 1 Mac class · 3 device families
RTX 3080~235Simulate output on RTX 3080RTX 2080 Ti~191Simulate output on RTX 2080 TiRTX 3070 Ti~188Simulate output on RTX 3070 TiM3*8 GB~34Simulate output on iMac 24" M3 (8 GB)Orin Nano~32Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~23Simulate output on iPhone 1720 GB
fits · 17.9 GB free · up to 256K ctx
1 GPU
RX 7900 XT~229Simulate output on RX 7900 XT32–40 GB
fits · 29.9 GB free · up to 256K ctx
1 GPU · 3 Mac classes · 9 device families
RTX 5090~555Simulate output on RTX 5090M5 Max*36 GB~153Simulate output on Mac Studio M5 Max (36 GB)M632 GB~57Simulate output on iMac 24" M6 (32 GB)RTX Spark N1X~93Simulate output on NVIDIA RTX Spark N1X Laptop (32 GB)Orin~63Simulate output on NVIDIA Jetson AGX Orin 32GB
By VRAM tier · typical GPUs and Macs at BF16
Similar models
1| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Qwen3.5 0.8BAlibaba | — | 0.9 GB | ~753 | same size |
Specifications
- Publisher
- togethercomputer
- Parameters
- 873M
- Architecture
- Qwen3_5ForConditionalGeneration
- Context Length
- 262,144 tokens
- Vocabulary Size
- 248,320
- Release Date
- 2026-09-25
Frequently Asked Questions
How much VRAM does Tev1 0.8B Experimental need?
Tev1 0.8B Experimental requires 2.1 GB of VRAM at BF16. Full 262K context adds up to 1.6 GB (3.7 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 873M × 16 bits ÷ 8 = 1.7 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 ≈ 2 GB (at full 262K context)
VRAM usage by quantization
Can I run Tev1 0.8B Experimental on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 158 other Macs can run Tev1 0.8B Experimental. Apple Silicon uses unified memory, so the model shares RAM with the system. At BF16 you need at least 2.1 GB of usable unified memory (RAM minus macOS overhead).
Can I run Tev1 0.8B Experimental locally?
Yes — Tev1 0.8B Experimental can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Tev1 0.8B Experimental?
At BF16, Tev1 0.8B Experimental can reach ~2286 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~312 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 ÷ 2.1 × 0.65 = ~2476 tok/s
Estimated speed at BF16 (2.1 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Tev1 0.8B Experimental?
At BF16, the download is about 1.75 GB.
Which GPUs can run Tev1 0.8B Experimental?
114 consumer GPUs can run Tev1 0.8B Experimental at BF16. Top options include AMD Radeon RX 5600 XT, AMD Radeon RX 5700 XT, AMD Radeon RX 580 8GB. 114 GPUs have plenty of headroom for comfortable inference.
Which devices can run Tev1 0.8B Experimental?
185 devices with unified memory can run Tev1 0.8B Experimental at BF16 (2.1 GB), including AMD Ryzen 7 8845HS (Hawk Point) Laptop, AMD Ryzen AI 7 350 (Krackan Point) Laptop, AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10. Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.