Hardware › A100 40GB local LLM benchmarks: 19 measured configs
A100 40GB local LLM benchmarks: 19 measured configs
19 measured local-LLM configs on NVIDIA A100 40GB; fastest 14B+ model Qwen3.6-35B-A3B at 124 tok/s. Official specs, every number linked to its source.
Snapshot 2026-10-15 · 394 configs on AI boxes and 423 on GPU cards · 123 models. Every number is generated from the tracker snapshot and linked to where it was measured; nothing here is typed by hand.
Quick answers
- What is the fastest model on A100 40GB?
- Qwen3.6-35B-A3B at 124 tok/s single-stream (UD-Q4_K_M, llama.cpp/cuda), source last checked 2026-10-14. Models under 14B parameters are left out of this answer. source
- How many configs does TokenMark track for A100 40GB?
- 19 single-stream configs across 6 models, each linked to the README or model card it came from.
Measured single-stream configs on NVIDIA A100 40GB
| Model | Quant | Backend | Decode | Context | Trust | Source |
|---|---|---|---|---|---|---|
| Qwen3.6-35B-A3B | UD-Q4_K_M | llama.cpp/cuda | 124 tok/s | 8192 | Verified (ran it) | source |
| Qwen3.5-35B-A3B | Q4_K_M | llama.cpp/cuda | 122.8 tok/s | 8192 | Verified (ran it) | source |
| Qwen3.6-35B-A3B | UD-Q4_K_XL | llama.cpp/cuda | 122.1 tok/s | 8192 | Verified (ran it) | source |
| Qwen3.6-35B-A3B | UD-Q5_K_M | llama.cpp/cuda | 120.9 tok/s | 8192 | Verified (ran it) | source |
| Qwen3.6-35B-A3B | UD-Q6_K | llama.cpp/cuda | 118.6 tok/s | 8192 | Verified (ran it) | source |
| gemma-4-26B-A4B-it | UD-Q8_K_XL | llama.cpp/cuda | 99.9 tok/s | 8192 | Verified (ran it) | source |
| Qwen3.8-27B | FP8 | vllm (speculative decode, mtp-4) | 81.2 tok/s | 32768 | Verified (ran it) | source |
| Qwen3.8-27B | FP8 | vllm (speculative decode, mtp-2) | 76.5 tok/s | 32768 | Verified (ran it) | source |
| Qwen3.8-27B | INT4 | vllm (speculative decode, mtp-4) | 72.2 tok/s | 65536 | Verified (ran it) | source |
| Qwen3.8-27B | INT4 | vllm (speculative decode, mtp-2) | 69 tok/s | 65536 | Verified (ran it) | source |
| Qwen3.8-27B | INT4 | vllm | 62.3 tok/s (prefill 3142.5) | 65536 | Verified (ran it) | source |
| Qwen3.8-27B | UD-Q4_K_M | llama.cpp/cuda | 45.5 tok/s | 8192 | Verified (ran it) | source |
| Qwen3.8-27B | UD-Q4_K_XL | llama.cpp/cuda | 43.5 tok/s | 8192 | Verified (ran it) | source |
| Qwen3.6-27B | Q4_K_M | llama.cpp/cuda | 42.4 tok/s | 8192 | Verified (ran it) | source |
| Qwen3.5-27B | Q4_K_M | llama.cpp/cuda | 42.4 tok/s | 8192 | Verified (ran it) | source |
| Qwen3.8-27B | FP8 | vllm | 41.2 tok/s (prefill 2830.4) | 65536 | Verified (ran it) | source |
| Qwen3.8-27B | UD-Q6_K_XL | llama.cpp/cuda | 34.7 tok/s | 8192 | Verified (ran it) | source |
| Qwen3.8-27B | Q8_0 | llama.cpp/cuda | 34.2 tok/s | 8192 | Verified (ran it) | source |
| Qwen3.6-27B | Q8_0 | llama.cpp/cuda | 34.1 tok/s | 8192 | Verified (ran it) | source |
More: all configs · hardware · news · contributors · methodology · llms.txt · llms-full.txt · API (OpenAPI) · JSON snapshot. Built by Altronis, private on-prem AI, Singapore.