AI Silicon Tracker
Every generation, what changed, and which AI numbers are actually real. Pick a company to follow one lineage, or compare all of them below.
7 of 12 companies here design AI silicon you cannot buy.
No company matches that.
-
Amazon
AMZN
Trainium and Inferentia — the largest cloud-captive alternative to NVIDIA
Captive
-
AMD
AMD
Instinct accelerators on ROCm — the open alternative to CUDA
Merchant
-
Apple
AAPL
The Neural Engine — an on-device NPU at consumer scale
Captive
9 chips
-
Broadcom
AVGO
Custom hyperscaler AI ASICs, and the Ethernet silicon that networks AI clusters
Custom
-
Google
GOOGL
The TPU — the leading non-NVIDIA AI training and inference accelerator
Captive
6 chips
-
Intel
INTC
The incumbent x86 server CPU vendor rebuilding both an accelerator line and a US foundry
Merchant
-
MediaTek
2454
Dimensity NPU — high-volume edge AI across mainstream Android
Merchant
3 chips
-
Microsoft
MSFT
Maia accelerators purpose-built for Azure inference
Captive
-
NVIDIA
NVDA
The merchant AI GPU standard — Tensor Core GPUs plus CUDA
Merchant
-
Qualcomm
QCOM
Hexagon NPU — on-device AI in Snapdragon, now scaled to rack-level inference
Merchant
5 chips
-
Samsung
005930
Exynos NPUs, HBM memory, and being a leading-edge foundry that fabricates its rivals’ AI chips
Mixed
-
Tesla
TSLA
The in-car FSD inference computer, and Dojo as a cautionary tale in custom training hardware
Captive
Every chip, all vendors
Apple A-series
| Chip | Announced | Node | Accelerator | What the vendor claims | Precision |
|---|---|---|---|---|---|
Apple |
Sep 2026 | 2nm | Dual 16-core Neural Engine (32 cores total) | Relative only2x the compute power to run on-device AI models | Not applicable — no absolute figure published |
Apple |
Sep 2025 | Third-generation 3nm | Neural Engine plus per-GPU-core Neural Accelerators | None publishedNo throughput figure published | Not applicable |
Apple |
Sep 2025 | Not stated by Apple ? | 16-core Neural Engine plus per-GPU-core Neural Accelerators | Relative onlyUp to 3x the peak GPU compute over the previous generation | Not applicable — and this is a GPU figure, not a Neural Engine one |
Apple |
Sep 2024 | Second-generation 3nm | 16-core Neural Engine | Relative onlyRuns machine learning models up to 2x faster than A16 Bionic | Not applicable — no absolute figure published |
Apple |
Sep 2024 | Second-generation 3nm | 16-core Neural Engine | Relative only17 per cent increase in total system memory bandwidth | Not applicable — no throughput figure published |
Apple |
Sep 2023 | 3nm | 16-core Neural Engine | Relative onlyNeural Engine up to 2x faster | Not applicable — no absolute figure published |
Apple |
Sep 2022 | Not stated by Apple ? | 16-core Neural Engine | AbsoluteNearly 17 trillion operations per second | Not stated by Apple |
Apple |
Sep 2021 | 5nm | 16-core Neural Engine | Absolute15.8 trillion operations per second | Not stated by Apple |
Apple |
Sep 2020 | 5nm | 16-core Neural Engine | Absolute11 trillion operations per second | Not stated by Apple |
Google Tensor
| Chip | Announced | Node | Accelerator | What the vendor claims | Precision |
|---|---|---|---|---|---|
| Aug 2026 | 3nm (TSMC) | Tensor Processing Unit (TPU), co-designed with Google DeepMind | Relative only50% more TPU compute than Tensor G5; on-device AI tasks up to 3.5x faster using up to 3.5x less energy | Not stated | |
| Aug 2025 | 3nm (TSMC) | Tensor Processing Unit (TPU) | Relative onlyUp to 60% more powerful TPU than Tensor G4; on-device Gemini Nano 2.6x faster and 2x more efficient; context window 12,000 to 32,000 tokens | Not stated | |
| Aug 2024 | Not stated by Google ? | Tensor Processing Unit (TPU) | None publishedNo TPU figure published | Not applicable | |
| Oct 2023 | Not stated by Google ? | Tensor Processing Unit (TPU) | Relative onlyRuns more than twice as many on-device models as Pixel 6; on-device generative AI 150x more complex than Pixel 7’s largest model | Not applicable | |
| Oct 2022 | Not stated by Google ? | Tensor Processing Unit (TPU) | None publishedNo quantified claim published | Not applicable | |
| Aug 2021 | Not stated by Google ? | Tensor Processing Unit (TPU) | None publishedNo figure published — qualitative only | Not applicable |
MediaTek Dimensity
| Chip | Announced | Node | Accelerator | What the vendor claims | Precision |
|---|---|---|---|---|---|
MediaTek |
Sep 2025 | TSMC N3P | NPU 990 with Generative AI Engine 2.0, plus a separate compute-in-memory always-on NPU | Relative only2x compute; 100% faster 3B-parameter LLM output; 128K-token on-device context; BitNet 1.58-bit support | Not stated. The widely cited 100 TOPS figure traces to Counterpoint Research and appears in no MediaTek-published material |
MediaTek |
Oct 2024 | Second-generation TSMC 3nm | NPU 890 | Relative onlyUp to 80% faster LLM prompt performance and up to 35% more power efficient than Dimensity 9300 | Not stated |
MediaTek |
Nov 2023 | Third-generation TSMC 4nm | APU 790 | Relative only8x faster than the previous generation for Transformer models; supports LLMs at 1B, 7B and 13B parameters, scalable to 33B | Doubling claimed for integer and floating point, but no bit width stated |
Snapdragon 8
| Chip | Announced | Node | Accelerator | What the vendor claims | Precision |
|---|---|---|---|---|---|
Qualcomm |
Sep 2025 | 3nm | Hexagon NPU with Fused AI Accelerator | Relative only37% faster Hexagon NPU, 16% better performance per watt | Not stated as a throughput precision, but the supported set widens to INT2, INT4, INT8, INT16, FP8, FP16 |
Qualcomm |
Oct 2024 | 3nm | Hexagon NPU | Relative only45% faster Hexagon NPU, 45% better performance per watt | Not stated. INT4, INT8, INT16, FP16 |
Qualcomm |
Oct 2023 | 4nm | Hexagon NPU | Relative onlyUp to 98% faster Hexagon NPU; on-device models up to 10B parameters at up to 20 tokens/sec (Llama 2 7B) | Not stated. INT4, INT8, INT16, FP16 |
Qualcomm |
Nov 2022 | 4nm | Hexagon NPU | Relative onlyUp to 4.35x faster AI performance than its predecessor | Not stated. Adds INT4 alongside INT8, INT16, FP16 |
Qualcomm |
Nov 2021 | 4nm | 7th Gen AI Engine with Hexagon processor | Relative onlyAI Engine up to 4x faster than predecessor | Not stated. Supports INT8, INT16, FP16 |
AI capability figures are reproduced exactly as each vendor published them. Most vendors publish no absolute throughput figure, and none of the mobile vendors state the numeric precision behind the figures they do give — which makes cross-vendor TOPS comparisons unreliable. Where a node is marked with a question mark, the vendor has not stated it.