GLM 5.3 Flash vs Muse Spark 1.3 Contributor, head to head

Side A

Side B

Overall

Muse Spark 1.3 Contributor is clearly better. It wins 63 of the 69 test cases where they differ.

Learn to build evals like this in the AI Engineering Academy

Task by task

A score turns orange only when that model wins clearly more of the task's test cases. Under each score, the average time per test case on the machine it ran on. Muse Spark 1.3 Contributor spends 3.2× as many thinking tokens per test case.

Thinking tokensAnswer tokens

TaskAGLM 5.3 FlashBMuse Spark 1.3 ContributorGLM 5.3 FlashMuse Spark 1.3 Contributor

Overall

3555.5s

2.8K8.7K

661m 42s

Agents

Runs multi-step jobs through tools, within a written policy

1037.0s

60714.7K

573m 16s

Structured Extraction

Turns logs, diffs and records into exact JSON

4847.2s

2.7K4.2K

9037.9s

Coding

Writes code and SQL that pass hidden tests

541m 52s

5.3K10.8K

482m 1s

Grounded Answers

Answers from the given documents with sources, never guesses

3756.8s

2.9K6.1K

811m 1s

Instruction Following

Applies every rule, including ones changed mid-conversation

4340.8s

2K8.7K

841m 18s

Business Logic

Computes billing, scheduling and pricing rules exactly

331m 18s

5.8K6K

761m 7s

Security & permissions

Ignores planted instructions, still does the allowed work

5525.2s

9638.2K

761m 59s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

027.0s

56317.2K

173m 34s

What you'd run

The build behind each score. Orange marks the faster, smaller, cheaper or longer-context side.

SpecAGLM 5.3 FlashBMuse Spark 1.3 Contributor

Speed

60 tok/s

107 tok/s

Machine

Hosted

Memory at 32K

—

Context

Cost per run

$0.19

$0.41

Cost per case

$0.0013

$0.0027

Time per case

55.5s

1m 42s

Quantization

FP8

—

Harness

openrouter

The race

No lead changes

Each model's running score as the test cases go by. The dashed line is a perfect run.

GLM 5.3 FlashMuse Spark 1.3 Contributor

Every test case

153 test cases

Each square is one test case, colored by the model that scored higher.

Agents010

Structured Extraction110

Coding55

Grounded Answers010

Instruction Following011

Business Logic09

Security & permissions07

Diagnosis & investigation01

6 won by GLM 5.3 Flash63 won by Muse Spark 1.3 Contributor84 tied