tinyanalysis

Leaderboard / SparkLLM (XHToken)

Spark-X2.6 1.7B

SparkLLM (XHToken) · Spark-X2 · released 2026-10-19 · data checked 2026-11-01

Spark-X2.6 1.7B is a 1.7B-parameter dense model from SparkLLM (XHToken) (China), released 2026-10-19 under the Apache-2.0 license. It scores 50.1 on the Small LM Index, #10 of 21 tiny models. It needs about 1.7 GB of memory at 4-bit with an 8K-token context.

Small LM Index
50.1
#10 of 21 scored tiny models
Likely 46.9–53.1 · from 4 of 6 core benchmarks
Parameters
1.7B
Dense
Price per 1M tokens
—
No hosted API on OpenRouter; self-host
Context window
1M
tokens

Benchmarks

Each bar shows this model's score. The tick marks the best tiny model on that benchmark. Sources: 11 reported by SparkLLM (XHToken). How we label sources.

GPQA Diamond Lab-reported Knowledge · GPQA Diamond56.1
Class best: Nanbeige4.2 3B 87.4
Humanity's Last Exam Lab-reported Reasoning8.7
Class best: Spark-X2.6 4B 17.9
AIME Lab-reported Math · AIME 202670.1
Class best: Spark-X2.6 4B 91.2
SWE-bench Verified Lab-reported Coding · Vendor harness59.8
Class best: Nanbeige4.2 3B 63.6
τ²-Bench Lab-reported Agentic85.3
Class best: MiniCPM5 2B 97.1
IFBench Lab-reported Instruction following65.5
Class best: Spark-X2.6 4B 75.8
IFEval Lab-reported Instruction following89.4
Class best: LFM3 1.7B 96.7
BFCL Lab-reported Agentic · BFCL v455.2
Class best: Spark-X2.6 4B 71.8
HMMT Lab-reported Math · HMMT Feb 202648.8
Class best: Nanbeige4.2 3B 82.8
AA-LCR Lab-reported Long context · AA-LCR v1.148.0
Class best: Spark-X2.6 4B 67.3
SciCode Lab-reported Coding20.1
Class best: Spark-X2.6 4B 36.8

Not reported: estimated

What the index model expects from this model's other scores. Not a measurement.

MMLU-Pro≈ 72
LiveCodeBench≈ 57

Specifications

Lab
SparkLLM (XHToken) (CN · China)
Architecture
Dense transformer
Total parameters
1.7B
Active parameters
All (dense)
Size class
Tiny
Context window
1M tokens
Input modalities
text
Reasoning
Hybrid (thinking on/off)
License
Apache-2.0
Openness
Open weights
Memory to run
~1.7 GB at 4-bit, 8K context (more sizes)
Hugging Face
2.7K downloads (30 days) · 72 likes
Weights on Hugging Face ↗

Compare with

Ranked in

Sources

Something wrong here? Report it.

Run it locally

Uses XHToken/Spark-X2.6-1.7B-GGUF (Q4_K_M). Swap the tag for another size: Q4_K_M, Q8_0.

Ollama
ollama run hf.co/XHToken/Spark-X2.6-1.7B-GGUF:Q4_K_M
llama.cpp
llama-server -hf XHToken/Spark-X2.6-1.7B-GGUF:Q4_K_M
Then open http://localhost:8080 for a chat window.
vLLM (GPU server)
vllm serve XHToken/Spark-X2.6-1.7B
Full-precision weights, for a server GPU.

Open in LM Studio

Memory needed: weights, context cache and runtime buffers.

Quantization4K context32K context256K context
16-bit F164.0 GB4.4 GB7.7 GB
8-bit Q8_02.4 GB2.8 GB6.1 GB
4-bit Q4_K_M1.7 GB2.1 GB5.4 GB

Check it against your GPU, Mac or phone →