Spark X2.6 1.7B
XHToken/Spark-X2.6-1.7B-GGUF
starter@v1Localllama.cppQ8_01M context2026-10-29 21:01 UTC4
3 of 153 test cases fully passed
106.9tok/s
Mean per test case, 106 tok/s over the run
10.8Ktokens
Answer plus reasoning
7.3Ktokens
68% of the output
19.0 GB
llama.cpp's figure for this run's 1M context and 4 slots
2.9M
1.3M in, 1.7M out
4h 21m
1m 43s per caseSum of request times
Not measured
No price recorded
Every test case
One square per test case. Select a square to read the answer.
0Agents0 of 21 passed
0Structured Extraction0 of 21 passed
0Coding0 of 21 passed
0Grounded Answers0 of 21 passed
24Instruction Following2 of 21 passed
0Business Logic0 of 21 passed
10Security & permissions1 of 21 passed
0Diagnosis & investigation0 of 6 passed
Speed and output per test case
One bar per test case. Hover a bar to see its numbers. Select it to read the answer.
Peak 115 tok/s
ReasoningAnswer
Test by test
Every answer with its reasoning and the checks it had to pass.
Reproduce this run
The prompts stay private so models can't train on them. The hashes pin the exact suite and config this run used.
llama.cpp
llama-server -hf XHToken/Spark-X2.6-1.7B-GGUF:Q8_0 --port 9931 -c 1048576 -ngl all --temp 1.0 --top-p 0.95 --top-k 0 --reasoning onPulls the GGUF from Hugging Face and serves an OpenAI-compatible endpoint on port 9931.
Configuration
Where the suite ran, and the token budget each task got.
- Local
- llama.cpp
- Q8_0
- -ngl all --temp 1.0 --top-p 0.95 --top-k 0 --reasoning on
- 1M
- Apple M5 Pro
- Apple M5 Pro, 18 cores
- 48 GB, shared with the GPU
- macOS 27.2, arm64
- Not set
- v1.0.0
- 2026-10-29 21:01 UTC
| Task | Reasoning | Max output per call |
|---|---|---|
| Agents | 4.1K | 12.3K |
| Structured Extraction | 4.1K | 12.3K |
| Coding | 4.1K | 12.3K |
| Grounded Answers | 4.1K | 12.3K |
| Instruction Following | 4.1K | 12.3K |
| Business Logic | 4.1K | 12.3K |
| Security & permissions | 4.1K | 12.3K |
| Diagnosis & investigation | 4.1K | 12.3K |