AMD-Pro-BenchCommunity

13,812 benchmarks in 32 GPU combinations documented

Ministral-3-8B-Instruct-2512 on R9700

148 results · 2 sources · 148 measurements

Explore all results

Grouped by reported model identity. Exact weights, quantizations and software can differ. How results are reviewed

Documented results

24 selected tests
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed2,048 input tokens · 16,384 existing KV tokens—
946.86 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; distribution unreported) (inferred)
Engine
llama.cpp
Backend
rocm-7-nightly
Harness
llama-bench
Prompt / output
2048 / ? tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed2,048 input tokens · 16,384 existing KV tokens—
1,569.29 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Fedora 42 — Linux 6.15.9-201.fc42.x86_64 (Sat Aug 2 11:37:34 UTC 2025)
Engine
llama.cpp
Backend
rocm6_4_4
Harness
llama-bench
Prompt / output
2048 / ? tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed32 output tokens · 16,384 existing KV tokens——
30.49 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; distribution unreported) (inferred)
Engine
llama.cpp
Backend
rocm-7-nightly
Harness
llama-bench
Prompt / output
? / 32 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed32 output tokens · 16,384 existing KV tokens——
31.15 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Fedora 42 — Linux 6.15.9-201.fc42.x86_64 (Sat Aug 2 11:37:34 UTC 2025)
Engine
llama.cpp
Backend
rocm6_4_4
Harness
llama-bench
Prompt / output
? / 32 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed2,048 input tokens · 32,768 existing KV tokens—
539.19 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; distribution unreported) (inferred)
Engine
llama.cpp
Backend
rocm-7-nightly
Harness
llama-bench
Prompt / output
2048 / ? tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed2,048 input tokens · 32,768 existing KV tokens—
961.48 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Fedora 42 — Linux 6.15.9-201.fc42.x86_64 (Sat Aug 2 11:37:34 UTC 2025)
Engine
llama.cpp
Backend
rocm6_4_4
Harness
llama-bench
Prompt / output
2048 / ? tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed32 output tokens · 32,768 existing KV tokens——
27.49 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; distribution unreported) (inferred)
Engine
llama.cpp
Backend
rocm-7-nightly
Harness
llama-bench
Prompt / output
? / 32 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed32 output tokens · 32,768 existing KV tokens——
28.09 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Fedora 42 — Linux 6.15.9-201.fc42.x86_64 (Sat Aug 2 11:37:34 UTC 2025)
Engine
llama.cpp
Backend
rocm6_4_4
Harness
llama-bench
Prompt / output
? / 32 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed512 input tokens · 0 existing KV tokens—
4,439.2 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; distribution unreported) (inferred)
Engine
llama.cpp
Backend
rocm-7-nightly
Harness
llama-bench
Prompt / output
512 / ? tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed2,048 input tokens · 65,536 existing KV tokens—
536.76 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Fedora 42 — Linux 6.15.9-201.fc42.x86_64 (Sat Aug 2 11:37:34 UTC 2025)
Engine
llama.cpp
Backend
rocm6_4_4
Harness
llama-bench
Prompt / output
2048 / ? tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed128 output tokens · 0 existing KV tokens——
34.52 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; distribution unreported) (inferred)
Engine
llama.cpp
Backend
rocm-7-nightly
Harness
llama-bench
Prompt / output
? / 128 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed32 output tokens · 65,536 existing KV tokens——
23.34 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Fedora 42 — Linux 6.15.9-201.fc42.x86_64 (Sat Aug 2 11:37:34 UTC 2025)
Engine
llama.cpp
Backend
rocm6_4_4
Harness
llama-bench
Prompt / output
? / 32 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed2,048 input tokens · 16,384 existing KV tokens—
411.98 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; distribution unreported) (inferred)
Engine
llama.cpp
Backend
rocm-7-nightly
Harness
llama-bench
Prompt / output
2048 / ? tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed512 input tokens · 0 existing KV tokens—
4,150.35 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Fedora 42 — Linux 6.15.9-201.fc42.x86_64 (Sat Aug 2 11:37:34 UTC 2025)
Engine
llama.cpp
Backend
rocm6_4_4
Harness
llama-bench
Prompt / output
512 / ? tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed32 output tokens · 16,384 existing KV tokens——
30.51 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; distribution unreported) (inferred)
Engine
llama.cpp
Backend
rocm-7-nightly
Harness
llama-bench
Prompt / output
? / 32 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed128 output tokens · 0 existing KV tokens——
35.32 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Fedora 42 — Linux 6.15.9-201.fc42.x86_64 (Sat Aug 2 11:37:34 UTC 2025)
Engine
llama.cpp
Backend
rocm6_4_4
Harness
llama-bench
Prompt / output
? / 128 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed2,048 input tokens · 32,768 existing KV tokens—
311.32 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; distribution unreported) (inferred)
Engine
llama.cpp
Backend
rocm-7-nightly
Harness
llama-bench
Prompt / output
2048 / ? tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed2,048 input tokens · 16,384 existing KV tokens—
1,782.24 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Fedora 42 — Linux 6.15.9-201.fc42.x86_64 (Sat Aug 2 11:37:34 UTC 2025)
Engine
llama.cpp
Backend
rocm7_2_2
Harness
llama-bench
Prompt / output
2048 / ? tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed32 output tokens · 32,768 existing KV tokens——
27.48 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; distribution unreported) (inferred)
Engine
llama.cpp
Backend
rocm-7-nightly
Harness
llama-bench
Prompt / output
? / 32 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed32 output tokens · 16,384 existing KV tokens——
30.78 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Fedora 42 — Linux 6.15.9-201.fc42.x86_64 (Sat Aug 2 11:37:34 UTC 2025)
Engine
llama.cpp
Backend
rocm7_2_2
Harness
llama-bench
Prompt / output
? / 32 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed512 input tokens · 0 existing KV tokens—
618.12 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; distribution unreported) (inferred)
Engine
llama.cpp
Backend
rocm-7-nightly
Harness
llama-bench
Prompt / output
512 / ? tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed2,048 input tokens · 32,768 existing KV tokens—
1,106.08 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Fedora 42 — Linux 6.15.9-201.fc42.x86_64 (Sat Aug 2 11:37:34 UTC 2025)
Engine
llama.cpp
Backend
rocm7_2_2
Harness
llama-bench
Prompt / output
2048 / ? tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed128 output tokens · 0 existing KV tokens——
34.52 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; distribution unreported) (inferred)
Engine
llama.cpp
Backend
rocm-7-nightly
Harness
llama-bench
Prompt / output
? / 128 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
Reportllama-bench		Ministral-3-8B-Instruct-2512BF16Model ref. ↗Token speed32 output tokens · 32,768 existing KV tokens——
27.69 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Fedora 42 — Linux 6.15.9-201.fc42.x86_64 (Sat Aug 2 11:37:34 UTC 2025)
Engine
llama.cpp
Backend
rocm7_2_2
Harness
llama-bench
Prompt / output
? / 32 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
Quantizations and software

Quantizations: BF16
Engines: llama.cpp

Explore all 148 results →

This is a selection across sources, not a ranking or a matched experiment. Reading benchmark reports

GPU benchmark pages: R9700