2 × W7800 LLM benchmarks
117 results · 360 measurements · 1 source
Device specifications
| Variant | Memory | Memory type | Bus | Bandwidth | FP16 matrix | INT8 matrix |
|---|---|---|---|---|---|---|
| W7800 48GB ↗ | 48 GB | GDDR6 | 384-bit | 864 GB/s | 90.4 TFLOPS | 90.4 TOPS |
| W7800 ↗ | 32 GB | GDDR6 | 256-bit | 576 GB/s | 90.4 TFLOPS | 90.4 TOPS |
Manufacturer specifications. Shared system RAM is not dedicated VRAM. Unreported values remain —; measured LLM results retain their reported configuration.
Selected reported tests
24 of 117 resultsSource-reported results, selected across reports by documentation quality. Match configurations before comparing values.
| View ReportCustom evaluation | Devstral-2-123B-Instruct-2512Conflicting evidence | HumanEval100 problems | 88 %Reported pass@1 · n = 100 | — | 7.05 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
|---|---|---|---|---|---|---|
| View ReportCustom evaluation | Devstral-Small-2-24B-Instruct-2512Conflicting evidence | HumanEval100 problems | 88 %Reported pass@1 · n = 100 | — | 31.57 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | gemma-4-26B-A4B-itConflicting evidence | HumanEval100 problems | 96 %Reported pass@1 · n = 100 | — | 71.18 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | GLM-4.5-AirConflicting evidence | HumanEval100 problems | 63 %Reported pass@1 · n = 100 | — | 37.33 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | GLM-4.7-FlashConflicting evidence | HumanEval100 problems | 90 %Reported pass@1 · n = 100 | — | 69.8 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | gpt-oss-120bConflicting evidence | HumanEval100 problems | 94 %Reported pass@1 · n = 100 | — | 74.78 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | Mistral-Small-4-119B-2603Conflicting evidence | HumanEval100 problems | 89 %Reported pass@1 · n = 100 | — | 56.51 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | Qwen3-Coder-30B-A3B-InstructConflicting evidence | HumanEval100 problems | 98 %Reported pass@1 · n = 100 | — | 94.08 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| 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 ReportCustom evaluation | Devstral-2-123B-Instruct-2512Conflicting evidence | Token speedsynth-bulk-dispatcher-impl | — | — | 7 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | Devstral-2-123B-Instruct-2512Conflicting evidence | Token speedsynth-horner-mpmath-derivative | — | — | 7.06 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | Devstral-2-123B-Instruct-2512Conflicting evidence | Token speedsynth-invoke-sign-artifacts-ps | — | — | 6.99 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | Devstral-2-123B-Instruct-2512Conflicting evidence | Token speedsynth-legacy-rename-toolchain | — | — | 7.04 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | Devstral-2-123B-Instruct-2512Conflicting evidence | Token speedsynth-nuget-multitarget-design | — | — | 7.05 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | Devstral-2-123B-Instruct-2512Conflicting evidence | Token speedsynth-parser-nunit-tests | — | — | 7.04 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | Devstral-2-123B-Instruct-2512Conflicting evidence | Token speedsynth-poincare-section-pytest | — | — | 6.96 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | Devstral-2-123B-Instruct-2512Conflicting evidence | Token speedsynth-range-expand-bug-fix | — | — | 7.1 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | Devstral-2-123B-Instruct-2512Conflicting evidence | Token speedsynth-relay-command-refactor | — | — | 7.04 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | Devstral-2-123B-Instruct-2512Conflicting evidence | Token speedsynth-resx-toolbar-button | — | — | 7.04 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | Devstral-2-123B-Instruct-2512Conflicting evidence | Token speedsynth-vanderpol-rk4-bugfix | — | — | 7.08 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | Devstral-2-123B-Instruct-2512Conflicting evidence | Token speedsynth-vanderpol-rk4-stepper-c | — | — | 7.02 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | Devstral-Small-2-24B-Instruct-2512Conflicting evidence | Token speedsynth-bulk-dispatcher-impl | — | — | 31.35 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | Devstral-Small-2-24B-Instruct-2512Conflicting evidence | Token speedsynth-horner-mpmath-derivative | — | — | 31.56 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
| View ReportCustom evaluation | Devstral-Small-2-24B-Instruct-2512Conflicting evidence | Token speedsynth-invoke-sign-artifacts-ps | — | — | 31.39 tok/sGeneration · mean · per request | 2 × W7800Layer split · 48 GB VRAM/GPUConfiguration
|
Models and software in this collection
Model families: Devstral 2, Devstral Small 2, GLM 4.5, GLM 4.7, GPT-OSS, Gemma 4, Mistral Small 4, Qwen3
Engines: Ollama
Methodology and evidence standards · Download the research dataset