LLM Reference

Spark X2.5 4B

Released
2026-08-31
Last refreshed
2026-09-01
Status
Researched today
Open sourceCommercial use: permittedCodingRAGAgentsLong contextJSON / Tool useAgentCoding

Spark X2.5 4B is available now for coding, rag, and agents with open-source and 1.05m context; evaluate it while provider pricing coverage matures.

Use it for

  • Teams evaluating coding, rag, and agents
  • Workloads that can use a 1.05m context window

Do not use it for

  • Cost-sensitive launches that need sourced token pricing
  • Vision or document-understanding workloads
  • Teams that need a tracked hosted API route today
Specifications
Released
2026-08-31
Context
1.05m
Parameters
4B
Architecture
Decoder Only
Specialization
agentic
Openness
Open source
License
Apache 2.0OSI-approvedCommercial use: permitted
Weights
Available
Code
Unknown
Training
Fine-tuned
Created by

Global communication through AI innovation

Hefei, China
Founded 1999
Website
Pricing

No tracked provider token pricing is available yet.

About

Spark-X2.5-4B is SparkLLM / XHToken (iFlytek) post-trained on-device general language model in the Spark-X2.5 series. Hugging Face card XHToken/Spark-X2.5-4B (Apache License 2.0 LICENSE blob, Copyright 2026 XHToken) documents a compact dense decoder (Spark2_5ForCausalLM) with hybrid attention: repeating 3 sliding-window layers plus 1 full-attention layer, native context up to 1,048,576 tokens (config max_position_embeddings), safetensors 4,112,079,360 parameters, thinking-mode chat template, tool-call parser spark25, and Llama-Factory fine-tuning. First-party GitHub README and HF card: pretrained ~20T tokens then SFT plus large-scale RL / MOPD from base XHToken/Spark-X2.5-4B-Base; 200+ languages; conversation, writing, translation, reasoning, coding, tool use, and agent harnesses (Codex, Claude Code, OpenClaw, Hermes).

Top use-case fit: coding, agents, and build tasks

Coding

Included by capability and metadata signals in the decision map.

RAG

Included by capability and metadata signals in the decision map.

Agents

Included by capability and metadata signals in the decision map.

Provider price ladder

No tracked provider token pricing is available for this model yet.

Capabilities

ReasoningJSON / Tool useFine-tuning

Benchmark peer barsfor Coding

No task-mapped benchmark peers are available for this model yet.

Migration checks

No linked migration route is available for this model yet.

API versions

spark-x2.5-4b