Xing4.0 29B-A4B has 31.215 billion total parameters. The selected IQ4_NL GGUF file is 20.104 GB.
Data snapshot: . 3 published tests in this catalog.
Model file and details
Provider
XingChen-AGI
Total parameters
31,215,031,088
Exact file size
20,104,012,544 bytes
Quantization / precision
IQ4_NL
Weight repository
XingChen-AGI/Xing4.0-29B-A4B-GGUF
Original model repository
XingChen-AGI/Xing4.0-29B-A4B
Weight revision
b8cd0a75f3d3dfa992bd8423ee22041b987eaf44
Model revision
baae3c3e813cad5f888f1f485cfff659c89076c5
Runtime notes
Published tests use very long contexts; this weight budget does not reproduce those settings.
Recorded release date
2026-09-16
Date evidence
Official source and GGUF repositories created on 16 September 2026; re-reviewed on 3 October after the source checkpoint reached the current Hugging Face trends.
These scores were compiled from the linked sources. These published scores were not independently tested here. Different prompts, precision, thinking budgets and tools can change results.
AIME 2026
90 / 100
Competition math from AIME 2026. Published answer accuracy.
Benchmark efficiency: 4.477 published score points per GB of selected weights. This ratio is not an inference-speed measure; check the test settings to see whether the score covers the selected quantization.
Test settings: Publisher-reported AIME 2026 accuracy for the BF16 source checkpoint, averaged over five runs with temperature 1.0, top_p 0.95 and a 131,072-token output limit.
The selected official IQ4_NL GGUF was not independently retested. Sampling, unusually large generation budgets and publisher harness details can affect comparison. Xing4.0 support is pending upstream and currently requires the publisher-linked llama.cpp port or another reviewed PR build.
Fixing real software issues with tools. Agent setup and external tools affect these scores. File size covers model weights only.
Benchmark efficiency: 3.731 published score points per GB of selected weights. This ratio is not an inference-speed measure; check the test settings to see whether the score covers the selected quantization.
Test settings: Publisher-reported SWE-bench Verified result for the BF16 source checkpoint using SWE-agent, temperature 1.0, top_p 0.95, repetition penalty 1.05 and a 210K context window.
The selected official IQ4_NL GGUF was not independently retested. Agent scaffold, tools, context and generation budget materially affect SWE-bench results; compare only with similarly configured runs.
Agentic terminal-use task success. Publisher-reported percentages are retained, but agent scaffold, tools, timeouts and reasoning budget can materially affect results.
Benchmark efficiency: 2.86 published score points per GB of selected weights. This ratio is not an inference-speed measure; check the test settings to see whether the score covers the selected quantization.
Test settings: Publisher-reported Terminal-Bench 2.1 average over three BF16-source runs in terminus-2, with temperature 0.8, top_p 0.95, repetition penalty 1.05, 64K max tokens and a 24-hour timeout.
The selected official IQ4_NL GGUF was not independently retested. Agent scaffold, tools, timeout and reasoning budget can materially affect Terminal-Bench results; compare cautiously.
Context and version evidence: The repository and file name identify the selected model. Revision hashes are shown above when recorded. Tested context length, full hardware configuration and run date are not consistently recorded in this snapshot. Only settings explicitly stated in each benchmark record are known.
The 20.104 GB figure covers model weights in this file. It excludes the context cache, runtime, operating system and any separate multimodal files. It is not a measured RAM requirement or a token-generation speed.
These complete configurations pass the initial weight-size check for the selected 20.10 GB file. Allow extra memory for context and software. Check the model’s runtime notes and your inference software before buying.
Listing evidence: 2026-09-25. Amazon.com base price visible in an owner-supplied screenshot reviewed 25 Sep 2026; capture time not supplied. Dated reference, not a live quote. Excludes card signup offers, coupons, tax and shipping.
Listing evidence: 2026-10-02. Apple US launch price for the base M5 Max Mac Studio, published 22 September 2026: USD 2,499. Dated manufacturer reference, not an Amazon offer; tax and shipping may differ.
Listing evidence: 2026-09-24. GMKtec, EVO-X2 / 64GB / 1TB, recorded 22 Sep 2026; exact ASIN capacity identified 24 Sep. This is a dated reference, not a current Amazon price; tax, shipping and bundle contents may differ.
Memory matches are estimates, not tested compatibility or speed. As an Amazon Associate I earn from qualifying purchases.
Published tests use very long contexts; this weight budget does not reproduce those settings. The selected weights use 20.104 GB; this excludes context cache, runtime buffers and other programs.
These are the first listed tiers that pass our reserve policy, not measured minimum requirements. System RAM and VRAM remain separate. Use the memory tool to change the reserve for your context and runner.