Advertise here
#398

Xing4.0 29b A4b

Xingchen AGI Release: 2026-10-17 Tested on: 2026-10-17 14:36 xingchen-agi/xing4.0-29b-a4b::medium
29B total (4B active)MoEOpen source
(medium) (none)

Summary

Xing4.0 29b A4b scores 3.7 on AI BENCHY and ranks #398. It has 7.3 reliability, a 26.1% pass rate, $0.000 total cost, and 371.69s average response time.

What makes Xing4.0 29b A4b unique: It stands out most in Domain specific, where it ranks #1, while Anti-AI Tricks is its weakest area at #8. Its total benchmark cost is unusually low for its score range.

Model facts

Researched on 2026-10-17

Reported
Parameters
29B total (4B active)
Architecture
MoE
Availability
Open source
License
Apache-2.0

The vendor reports 29B total and 4B active parameters, public Apache-2.0 weights, Transformers custom code and serving instructions. The native contract toggles enable_thinking; it does not declare distinct effort levels. Configure one enabled mode, using medium as the OpenRouter transport setting, plus thinking disabled, rather than publishing invented low/high differences. Phala is the current free-route provider and publishes training=false, retainsPrompts=false. Keep the exact :free execution ID; facts and fallback release-date keys use the base ID. Fallback date records this onboarding request date.

Score

3.7

Consistency

6.2

Total Output Tokens

107,980

Total Input Tokens

98,467

Input Price

$0.000 / 1M

Output Price

$0.000 / 1M

Cache Read Price

N/A

Cache Write Price

N/A

Cache prices apply to input tokens. Reads reuse cached prompts; writes store them and can cost extra. Output tokens use the output price.

Tests Correct

Wrong Tests: 22

Attempt pass rate: 26.1%

Flaky tests

11

Flaky tests had mixed outcomes across runs (at least one pass and one fail).

Response Time (avg)

371.69s

Response Time (max): 971.85s

Response Time (total): 8548.88s

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#398 Xing4.0 29b A4b

medium
Reached the allocated time limit (600 seconds) without receiving showcase output.
Cost
$0.000
Time
600.0s
Tokens
0 tok

Charts

Choose the first model, then click a second model to open a side-by-side page.

Total Output Tokens

Score vs Total Output Tokens

Quick Compare

Category Breakdown

Category Score Consistency Tests Correct
Agentic 3.0 10.0
Anti-AI Tricks 3.8 1.6
Coding 4.2 5.1
Combined 3.0 10.0
Data parsing and extraction 2.9 5.9
Domain specific 3.6 7.2
General Intelligence 3.2 9.9
Instructions following 4.3 6.8
Puzzle Solving 2.9 4.4
Tool Calling 10.0 10.0
Trivia 3.0 10.0

Compared models