Grouped by reported model identity. Exact weights, quantizations and software can differ. How results are reviewed
Documented results
24 selected tests| View ReportCustom evaluation | Qwen3-Coder-NextConflicting evidence | HumanEval100 problems | 96 %Reported pass@1 · n = 100 | — | 67.21 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
|---|---|---|---|---|---|---|
| View ReportLocalMaxxing | Qwen3-Coder-Next | Token speed | — | 433.38 tok/sPrefill · median · unknown | 46.97 tok/sGeneration · median · unknown | 1 × Ryzen AI Max+ 395Single GPU · Unified memoryConfiguration
|
| View ReportLocalMaxxing | Qwen3-Coder-Next | Token speed | — | 1,377.01 tok/sPrefill · unknown | 68.87 tok/sGeneration · peak · unknown | 3 × R9700Multi-GPU · split unknown · 32 GB VRAM/GPUConfiguration
|
| View ReportGeneric LLM | Qwen3-Coder-NextConflicting evidence | Lemonade benchmark2026.39.1 · code-debug | 0.4 tok/sThroughput · mean · unknown | — | — | 1 × R9700Single GPU · 32 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Qwen3-Coder-Next | Token speed | — | 496.38 tok/sPrefill · mean · per request | — | 1 × Ryzen AI Max+ 395Single GPU · Unified memory · 128 GB RAMConfiguration
|
| View ReportCustom evaluation | Qwen3-Coder-NextConflicting evidence | Token speedsynth-bulk-dispatcher-impl | — | — | 66.86 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportLocalMaxxing | Qwen3-Coder-Next | Token speed | — | 1,123.63 tok/sPrefill · unknown | 62.54 tok/sGeneration · peak · unknown | 3 × R9700Multi-GPU · split unknown · 32 GB VRAM/GPUConfiguration
|
| View ReportGeneric LLM | Qwen3-Coder-NextConflicting evidence | Lemonade benchmark2026.39.1 · code-debug | 0.5 tok/sThroughput · mean · unknown | — | — | 1 × R9700Single GPU · 32 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Qwen3-Coder-Next | Token speed | — | — | 31.92 tok/sGeneration · mean · per request | 1 × Ryzen AI Max+ 395Single GPU · Unified memory · 128 GB RAMConfiguration
|
| View ReportCustom evaluation | Qwen3-Coder-NextConflicting evidence | Token speedsynth-horner-mpmath-derivative | — | — | 67.61 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportLocalMaxxing | Qwen3-Coder-Next | Token speed | — | 1,081.77 tok/sPrefill · unknown | 48.18 tok/sGeneration · peak · unknown | 3 × R9700Multi-GPU · split unknown · 32 GB VRAM/GPUConfiguration
|
| View ReportGeneric LLM | Qwen3-Coder-NextConflicting evidence | Lemonade benchmark2026.39.1 · code-short | 50.4 tok/sThroughput · mean · unknown | — | — | 1 × R9700Single GPU · 32 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Qwen3-Coder-Next | Token speed | — | 481.4 tok/sPrefill · mean · per request | — | 1 × Ryzen AI Max+ 395Single GPU · Unified memory · 128 GB RAMConfiguration
|
| View ReportCustom evaluation | Qwen3-Coder-NextConflicting evidence | Token speedsynth-invoke-sign-artifacts-ps | — | — | 66.82 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportLocalMaxxing | Qwen3-Coder-Next | Token speed | — | 1,351.93 tok/sPrefill · unknown | 66.4 tok/sGeneration · peak · unknown | 3 × R9700Multi-GPU · split unknown · 32 GB VRAM/GPUConfiguration
|
| View ReportGeneric LLM | Qwen3-Coder-NextConflicting evidence | Lemonade benchmark2026.39.1 · code-short | 55.1 tok/sThroughput · mean · unknown | — | — | 1 × R9700Single GPU · 32 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Qwen3-Coder-Next | Token speed | — | — | 31.9 tok/sGeneration · mean · per request | 1 × Ryzen AI Max+ 395Single GPU · Unified memory · 128 GB RAMConfiguration
|
| View ReportCustom evaluation | Qwen3-Coder-NextConflicting evidence | Token speedsynth-legacy-rename-toolchain | — | — | 67.24 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportLocalMaxxing | Qwen3-Coder-Next | Token speed | — | — | 80.82 tok/sGeneration · peak · unknown | 2 × R9700Multi-GPU · split unknown · 32 GB VRAM/GPUConfiguration
|
| View ReportGeneric LLM | Qwen3-Coder-NextConflicting evidence | Lemonade benchmark2026.39.1 · code-explain | 50.1 tok/sThroughput · mean · unknown | — | — | 1 × R9700Single GPU · 32 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Qwen3-Coder-Next | Token speed | — | 487.41 tok/sPrefill · mean · per request | — | 1 × Ryzen AI Max+ 395Single GPU · Unified memory · 128 GB RAMConfiguration
|
| View ReportCustom evaluation | Qwen3-Coder-NextConflicting evidence | Token speedsynth-nuget-multitarget-design | — | — | 67.41 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportLocalMaxxing | Qwen3-Coder-Next | Token speed | — | — | 80.4 tok/sGeneration · peak · unknown | 3 × R9700Multi-GPU · split unknown · 32 GB VRAM/GPUConfiguration
|
| View ReportGeneric LLM | Qwen3-Coder-NextConflicting evidence | Lemonade benchmark2026.39.1 · code-explain | 55.7 tok/sThroughput · mean · unknown | — | — | 1 × R9700Single GPU · 32 GB VRAM/GPUConfiguration
|
Quantizations and software
Quantizations: FP8, MXFP4_MOE, Q4_K_M, Q5_K_M, UD-Q5_K_M, UD-Q6_K_XL
Engines: Lemonade / llama.cpp, llama.cpp, Ollama, vllm
This is a selection across sources, not a ranking or a matched experiment. Reading benchmark reports
GPU benchmark pages: W7800 · R9700 · Ryzen AI Max+ 395