Benchmark result
Cyber-Tiel-Coder-35B-A3B-UD-Q4_K_XL on Intel Graphics — 17.1 tok/s
Measured with llama.cpp manual on October 3, 2026.
Support llm-bench.io
llm-bench.io is free. If it helped you, I'd appreciate your support to cover the infrastructure costs.
What the model built
Open full screen →The coding scenario asks for a playable game in a single HTML file. This is exactly what the model returned, unedited; the frame only adds a line that tells this page how tall it is.
How this run compares
1st fastest of 6 runs of this model on Intel Graphics · median 16.7 tok/s.
sort
- 117.1tok/s79.4llama.cpp · Q4_Kthis run
- 217.0tok/s84.4llama.cpp · Q4_K
- 316.9tok/s79.4llama.cpp · Q4_K
- 416.5tok/s85.4llama.cpp · Q4_K
- 515.8tok/s82.4llama.cpp · Q4_K
- 615.6tok/s81.5llama.cpp · Q4_K
Reproduce this run
toolllama.cpp manual
modelCyber-Tiel-Coder-35B-A3B-UD-Q4_K_XL (Q4_K)
context260,000
thinkingon
think limitnone
samplingtemperature 0.8999999761581421 · top_p 0.949999988079071 · top_k 20 · min_p 0 · presence_penalty 0 · repeat_penalty 1
clientv0.4.87+102
Set those in llama.cpp, then:
llm-benchmark benchmark --model "Cyber-Tiel-Coder-35B-A3B-UD-Q4_K_XL" --tool "llama.cpp"The client prompts for context length and thinking mode, and for the KV cache dtype on Unsloth Studio. Sampling is left at the model default — the values above are what the tool reported using, not overrides the client sent.