Consumer GPUAMD

AMD Radeon RX 7800 XT 16GB

Curated Aggregate·2025-05-233 workloads · 4 records
MyAI Rating8.1tok/s · Phi-3 Mini

VRAM

16 GB

TDP

263 W

MSRP

$499

Perf/W

0.24 tok/s/W

Cost/1K tok

$0.08/M

Tested

2025-05-23

Quick answer

How fast is AMD Radeon RX 7800 XT 16GB for local AI workloads?

AMD Radeon RX 7800 XT 16GB hits 132.0 tok/s on Phi-3 Mini, its strongest benchmarked workload (batch 1, 4096-token context, 16GB VRAM, 263W TDP). It has 4 records across 3 workloads in our database, with a MyAI Rating of 8.1/10. Llama 3 70B Q4 needs ~40GB VRAM, so check the VRAM column before assuming feasibility.

Source: MyAIHardware benchmark database (bench-x3-rx7800xt-phi3)As of 2024-08-30

Verdict

AMD Radeon RX 7800 XT 16GB with 16GB VRAM at 263W TDP, scored across 3 workloads with 4 benchmark records.

Best workload

Phi-3 Mini

132 tok/s

Quantization

Q4_K_M

4K context · batch 1

LLM Inference Performance

03570105140Llama 38B Q4Mistral 7BPhi-3 Mini

Benchmarks (3 workloads)

WorkloadScoreQuantContextσStatusTested
Llama 3 8B Q4

llm

64.0tok/sQ4_K_M4K, Curated Aggregate2025-05-23
Mistral 7B

llm

72.0tok/sQ4_K_M4K, Curated Aggregate2025-04-25
Phi-3 Mini

llm

132.0tok/sQ4_K_M4K, Curated Aggregate2024-08-30

MyAI Score

Enthusiast
8.1/10

AMD Radeon RX 7800 XT 16GB clears a 8.1/10 based on workload-normalized throughput, memory headroom, efficiency, value, trust, and coverage.

Throughput
286
Capability
126
Efficiency
46
Value
34
Trust
43
Coverage
20
Composite benchmark555 / 1000

Workload Fit

What models fit this 16GB card at different quantization levels.

Q4
Q8
FP16
7-8B
Excellent
Excellent
Tight
13-14B
Excellent
Good
Won't fit
32B
Tight
Won't fit
Won't fit
70B
Won't fit
Won't fit
Won't fit
Top Benchmarks
Phi-3 Mini132 tok/s
Mistral 7B72 tok/s
Llama 3 8B Q464 tok/s

Source

Phi-3 Mini on RDNA 3.

View sourceHow we benchmark →

Public Trust Layer

Trust score

6/10

MyAI rating

8.1

Runs

1

Freshness

Stale

Source-linked row with explicit verification status.

Tested on 2024-08-30; 727 days old.

Open primary source

Best place to start

Where to buy

Most buyers

Retailer we'd check first

Amazon

For mainstream AI hardware, Amazon usually updates street pricing, seller availability, and shipping speed faster than most comparison sites.

View current Amazon listing
  • +Use $499 as your price anchor unless the part is clearly supply-constrained or newly launched.
  • +Check the exact cooler, board partner, or memory configuration before buying. The silicon may match, but noise and thermals do not.
  • +If the Amazon price looks inflated, wait or compare against recent street pricing rather than paying a panic premium.

This is a strong buy when the listing stays close to reference pricing and matches the workload you actually run.

Affiliate note: this button opens the mapped Amazon product listing for this device. We may earn from qualifying purchases.