AI chip · NVIDIA
NVIDIA T4 Tensor Core GPU
Turing-generation 70W PCIe inference GPU, 16 GB GDDR6. Dominant cloud inference SKU 2019-2022; still heavily deployed on AWS EC2 g4 + GCP T4 instances.
Market position
The mainstream data-center inference GPU of 2019-2022. Sub-$3k, 70 W, single-slot. Every hyperscaler put fleets of these into VDI, ML inference, and video-transcoding jobs. A10 succeeded it in 2021 for higher throughput; L4 succeeded it in 2023 for Ada-generation efficiency.
DEPLOY editorial. What the vendor PDF cannot tell you.
Physical-AI cross-link
Derived from claims already on record (vendor first-party TDP, throughput, price, and the reference-design cluster registry). The arithmetic is shown per row so it can be audited. No estimated inputs.
- Perf per watt
- 0.93 FP16 TFLOPS/W65 TFLOPS ÷ 70 W = 0.93
What fits in 16 GB
Weights-only footprint for public open-weight LLMs. Pure arithmetic: params × bytes/param. Excludes KV cache and activation memory; add ~10-30% headroom for real serving. A model that does not fit at FP16 may still fit at INT8 or INT4 with quality trade-offs. Not a benchmark.
| Model | Params | FP16 | INT8 | INT4 |
|---|---|---|---|---|
| Llama 3.1 8B dense | 8 B | 16 GB ✓ | 8 GB ✓ | 4 GB ✓ |
| Llama 3.1 70B dense | 70 B | 140 GB ✗ | 70 GB ✗ | 35 GB ✗ |
| Llama 3.1 405B dense | 405 B | 810 GB ✗ | 405 GB ✗ | 203 GB ✗ |
| Llama 3.3 70B dense | 70 B | 140 GB ✗ | 70 GB ✗ | 35 GB ✗ |
| DeepSeek V3 MoE (671B total, 37B active per token) | 671 B | 1342 GB ✗ | 671 GB ✗ | 336 GB ✗ |
| DeepSeek R1 MoE (671B total, 37B active per token) | 671 B | 1342 GB ✗ | 671 GB ✗ | 336 GB ✗ |
| Qwen 2.5 7B dense | 7 B | 14 GB ✓ | 7 GB ✓ | 4 GB ✓ |
| Qwen 2.5 72B dense | 72 B | 144 GB ✗ | 72 GB ✗ | 36 GB ✗ |
| Mixtral 8x7B MoE (46.7B total, 12.9B active per token) | 46.7 B | 93 GB ✗ | 47 GB ✗ | 23 GB ✗ |
| Mixtral 8x22B MoE (141B total, 39B active per token) | 141 B | 282 GB ✗ | 141 GB ✗ | 71 GB ✗ |
| Gemma 2 27B dense | 27 B | 54 GB ✗ | 27 GB ✗ | 14 GB ✓ |
| Command R+ dense | 104 B | 208 GB ✗ | 104 GB ✗ | 52 GB ✗ |
| Kimi K2 MoE (1T total, 32B active per token) | 1000 B | 2000 GB ✗ | 1000 GB ✗ | 500 GB ✗ |
Math: FP16 = params × 2 bytes; INT8 = params × 1 byte; INT4 = params × 0.5 bytes. MoE models sum every expert (full weights on disk), not the per-token active subset.
Common questions
Answer-first, sourced. Every claim below traces to a specific field on this page or a cited datasheet. FAQPage schema is emitted so LLM crawlers can lift these verbatim.
How much does NVIDIA T4 Tensor Core GPU cost?
NVIDIA T4 Tensor Core GPU launch pricing is not on record in the DEPLOY registry. Vendor list prices are typically published in the vendor's own briefing rather than the datasheet.
How much memory does NVIDIA T4 Tensor Core GPU have?
16 GB of GDDR6 at 320 GB/s. Vendor datasheet, first-party.
How much power does one NVIDIA T4 Tensor Core GPU draw?
70 W TDP (thermal design power) per chip. System-level draw is higher: see the reference-design section for rack-level kW.
Which open-weight LLMs fit on one NVIDIA T4 Tensor Core GPU?
Of the 13 public open-weight LLMs DEPLOY tracks: 2 fit at FP16, 2 at INT8, 3 at INT4 (weights only, excludes KV cache). The full table is above with per-model math. Sparsity, offloading and multi-GPU serving change the picture; this row is single-chip weights-only.
Key facts
- Class
- Compute SoC (AI accelerator)
- Designer
- NVIDIA
- Safety-critical?
- No (data-center inference / training)
- Record as of
- 2026-08-25
Specifications (14 fields, click to expand)
Vendor datasheet figures (first-party). Dense throughput first; sparse (2:4) in parentheses where NVIDIA quotes it. The exhaustive spec sheet lives on the datasheet URL below the table: this row set covers what buyers actually compare on.
- Process node
- TSMC 12FFN
- Transistors
- 13.6 B
- Die size
- 545 mm²
- CUDA cores
- 2,560
- Tensor cores
- 320 (2nd gen (Turing))
- TDP
- 70 W
- Memory
- 16 GB GDDR6
- Memory bandwidth
- 320 GB/s
- PCIe
- Gen 3 x16
- FP16 (dense)
- 65 TFLOPS
- INT8 (dense)
- 130 TOPS
- Form factor
- PCIe (single-slot, low-profile)
- Announced
- 2018-09-13
- Released
- 2019-01-01
Source: vendor datasheet
Generation
Compare with
Sources
- NVIDIA T4Nvidia
Adoption rows appear as we verify chip-integration claims per model. Every row is a public claim the maker or a tier-1 source has stated; verification tier (verified / reported / inferred) is shown per row. See every chip on record for the full catalog or the NVIDIA record.