[whatmodel.sh](../../README.md) / Spark-X2.6 4B
# Spark-X2.6 4B
iFLYTEK (SparkLLM) · released 2026-10 · 4.112B params · apache-2.0
**score 67.2** ±1.0 · #111 of 732 · from 17 benchmark results across 1 source · † no independently run results verified
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks Spark-X2.6 4B among every model on that task or test, worst to best.
task vs the others rank counts tests
agent tasks ▁▁▃▆█▄▅●▄▄▂▁ #94 of 313 31% 7
coding ▁▁▁▃▅▇█▇●▄▃▁ #101 of 469 22% 4
chat & writing ▁▁▁▃▆▄▆█▆●▇▃ #100 of 402* 2.1% 1
math ▁▁▃▅█▆▄●▄▂▁▁ #95 of 431 18% 3
web search ▅·▅▅█·▃▃▃·▅● #1 of 15* 7.5% 2
agent tasks · #94 of 313 · 31% of its score
Beats [Grok 4.20 (0309 reasoning)](grok-4-20.md), [GPT-5.1](gpt-5-1.md), [o3-pro](o3-pro.md) here, though they rank higher overall. Trails [Hy3 preview (295B-A21B)](hy3-preview.md), [MiniMax-M2.7](minimax-m2-7.md), [Step 3.5 Flash (196B-A11B)](step-3-5-flash.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
SWE-bench Pro (split unspecified)… 54.3% ▂·▁▁▁▁▂▃●▇▂▁ beats 32/58 very hard · avg 9.8% 10% sharp · at its level
BrowseComp (maker) 53.1% ▂·▃▁▂▂▆●█▂▂▄ beats 22/46 hard · avg 21% 4.1% at its level
SWE-bench Verified (maker) 62.8% ▁▁▁▁▁▁▂▄●▆▁▁ beats 64/138 hard · avg 19% 3.9% sharp · at its level
SWE-bench Multilingual (maker) 64.7% ▁··▁▃▁▅●█▂▂▁ beats 28/65 hard · avg 25% 3.5% at its level
Humanity's Last Exam with tools (… 36.9% ▁·▁▂▂▁▄●▇█▄▅ beats 18/62 very hard · avg 11% 3.4% at its level
BFCL V4 (December 2025 archive) (… 71.8% ▁▁▁▂▆▆▅▆█▂▄● beats 108/114 medium · avg 40% 3.0% at its level
τ²-bench aggregate (maker) 86.9% ▁▁▁·▁▁▃▃▃▄●▁ beats 34/39 medium · avg 54% 2.6% typical
coding · #101 of 469 · 22% of its score
Beats [Grok 4.20 (0309 reasoning)](grok-4-20.md), [Grok 4.3](grok-4-3.md), [Muse Glimmer 30B](muse-glimmer-30b.md) here, though they rank higher overall. Trails [Llama 3.1 Nemotron Ultra 253B v1](llama-3-1-nemotron-ultra-253b.md), [Qwen3 235B-A22B Thinking 2507](qwen3-235b-a22b-thinking-2507.md), [o4-mini](o4-mini.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
SWE-bench Pro (split unspecified)… 54.3% ▂·▁▁▁▁▂▃●▇▂▁ beats 32/58 very hard · avg 9.8% 10% sharp · at its level
SciCode (level unspecified) (make… 36.8% ▁····▃▃▄▅●▆▂ beats 22/40 very hard · avg 14% 4.2% at its level
SWE-bench Verified (maker) 62.8% ▁▁▁▁▁▁▂▄●▆▁▁ beats 64/138 hard · avg 19% 3.9% sharp · at its level
SWE-bench Multilingual (maker) 64.7% ▁··▁▃▁▅●█▂▂▁ beats 28/65 hard · avg 25% 3.5% at its level
chat & writing · #100 of 402 · 2.1% of its score
Beats [LongCat 2.0 (1.6T-A48B)](longcat-2-0.md), [Inkling Small (266B-A12B)](inkling-small.md), [GPT-5.4 nano](gpt-5-4-nano.md) here, though they rank higher overall. Trails [Gemma 4 12B](gemma-4-12b.md), [Kimi K2 Instruct (0905)](kimi-k2-0905.md), [Gemma 4 26B-A4B](gemma-4-26b-a4b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
IFEval (aggregation unspecified) … 93.6% ▁▂▄▂▂▄▆██●▂▁ beats 135/145 easy · avg 83% 2.1% maxed out
math · #95 of 431 · 18% of its score
Beats [Claude Sonnet 4.6](claude-sonnet-4-6.md), [Claude Opus 4.5](claude-opus-4-5.md), [MiniMax-M3](minimax-m3.md) here, though they rank higher overall. Trails [Gemini 2.5 Flash](gemini-2-5-flash.md), [Granite 4.2 30B](granite-4-2-30b.md), [Qwen3-Next 80B-A3B Instruct](qwen3-next-80b-a3b-instruct.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
HMMT February 2026 (maker) 81.3% ▃▃▂▅·██●█▅▃▅ beats 22/35 hard · avg 32% 6.6% sharp
IMO-AnswerBench (maker) 75.2% ▁▁·▁▁▁▄●█▄▂▃ beats 15/44 hard · avg 32% 6.2% typical
AIME 2026 (maker) 91.2% ▁▁▁▂▂▂▆●▆▂·▁ beats 27/52 medium · avg 53% 5.4% sharp
web search · #1 of 15 · 7.5% of its score
test result vs the others in its run difficulty counts why
BrowseComp (maker) 53.1% ▂·▃▁▂▂▆●█▂▂▄ beats 22/46 hard · avg 21% 4.1% at its level
Humanity's Last Exam with tools (… 36.9% ▁·▁▂▂▁▄●▇█▄▅ beats 18/62 very hard · avg 11% 3.4% at its level
other tests (5)
test result vs the others in its run difficulty counts why
GPQA Diamond (maker) 77.1% ▁▁▁▁▂▄▅▅█●█▄ beats 190/322 hard · avg 38% 14% sharp · at its level
GPQA (subset unspecified) (maker) 69.9% ▂▁▂▂▄▃▆▃▅●▃▃ beats 77/106 hard · avg 37% 11% sharp · at its level
Humanity's Last Exam (maker) 17.9% ▁▁▁▄▆▆●██▆▁▃ beats 45/111 very hard · avg 3.8% 8.3% typical
LongBench v2 (maker) 53.3% ▂·▁·▂▃▁▄▄●█▄ beats 19/33 very hard · avg 32% 7.9% at its level
IFBench (aggregation unspecified)… 75.8% ▂▆▂▄▄▄▂▄▄▆●▇ beats 68/86 medium · avg 49% 3.0% typical
source results share of the score
Spark-X2.6 model cards (submitted / mak… 17 100% ▇▇▇▇▇▇▇▇▇▇▇▇
prior: pull toward the middle 0% ▇
weighted average = 67.2
Licences: Spark-X2.6 model cards ([Apache-2.0](../../licenses/spark-x26.txt))
## How to run it
### llama.cpp (recommended)
1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md)
```sh
# macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
# Windows
irm https://llama.app/install.ps1 | iex
```
2. Run Spark-X2.6 4B (Q4_K_M, needs ~3.5434 GB, from [huggingface.co/XHToken](https://huggingface.co/XHToken/Spark-X2.6-4B-GGUF/tree/0e8aa39a34063946ce07bbaaff392093b11b6829))
```sh
llama cli -hf XHToken/Spark-X2.6-4B-GGUF
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf XHToken/Spark-X2.6-4B-GGUF
```
Other versions: Q8_0 (~4.375 GB) `llama cli -hf XHToken/Spark-X2.6-4B-GGUF` · BF16 (~8.2299 GB) `llama cli -hf XHToken/Spark-X2.6-4B-GGUF`
### LM Studio
1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download)
2. Run Spark-X2.6 4B (Q4_K_M, needs ~3.5434 GB, from [huggingface.co/XHToken](https://huggingface.co/XHToken/Spark-X2.6-4B-GGUF/tree/0e8aa39a34063946ce07bbaaff392093b11b6829))
```sh
lms get https://huggingface.co/XHToken/Spark-X2.6-4B-GGUF -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: Q8_0 (~4.375 GB) `lms get https://huggingface.co/XHToken/Spark-X2.6-4B-GGUF -y` · BF16 (~8.2299 GB) `lms get https://huggingface.co/XHToken/Spark-X2.6-4B-GGUF -y`
### Ollama
1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download)
```sh
# Linux
curl -fsSL https://ollama.com/install.sh | sh
```
2. Run Spark-X2.6 4B (Q4_K_M, needs ~3.5434 GB, from [huggingface.co/XHToken](https://huggingface.co/XHToken/Spark-X2.6-4B-GGUF/tree/0e8aa39a34063946ce07bbaaff392093b11b6829))
```sh
ollama run hf.co/XHToken/Spark-X2.6-4B-GGUF
```
Other versions: Q8_0 (~4.375 GB) `ollama run hf.co/XHToken/Spark-X2.6-4B-GGUF` · BF16 (~8.2299 GB) `ollama run hf.co/XHToken/Spark-X2.6-4B-GGUF`
Model repository and documentation: [huggingface.co/XHToken](https://huggingface.co/XHToken/Spark-X2.6-4B)
[interactive version](https://whatmodel.sh/model/spark-x2-6-4b) · [JSON](https://whatmodel.sh/api?format=json&all=1)
