Beam 1.0 501B A23B vs Spark X2.6 4B, head to head

Side A

Side B

Overall

Beam 1.0 501B A23B is clearly better. It wins 82 of the 90 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. Spark X2.6 4B spends 2.0× as many thinking tokens per test case.

Thinking tokensAnswer tokens

TaskABeam 1.0 501B A23BBSpark X2.6 4BBeam 1.0 501B A23BSpark X2.6 4B

Overall

5717.2s

4K8.2K

151m 3s

Agents

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

5521.8s

4.8K11.5K

141m 32s

Structured Extraction

Turns logs, diffs and records into exact JSON

819.98s

3.4K5.3K

1039.5s

Coding

Writes code and SQL that pass hidden tests

4819.9s

4.5K9.1K

41m 8s

Grounded Answers

Answers from the given documents with sources, never guesses

6911.6s

2.7K6.6K

1050.2s

Instruction Following

Applies every rule, including ones changed mid-conversation

7820.9s

5.6K11.9K

371m 29s

Business Logic

Computes billing, scheduling and pricing rules exactly

5714.0s

4.2K4.9K

1036.3s

Security & permissions

Ignores planted instructions, still does the allowed work

5120.6s

3.2K7K

3154.3s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

1724.3s

3.1K11.6K

01m 30s

What you'd run

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

SpecABeam 1.0 501B A23BBSpark X2.6 4B

Speed

274 tok/s

136 tok/s

Machine

Hosted

RTX 5090, 32 GB

Memory at 32K

—

9.4 GB

Context

605K

Cost per run

$0.83

$1.61 est.

Cost per case

$0.0054

$0.011 est.

Time per case

17.2s

1m 3s

Quantization

FP8

BF16

Harness

openrouter

llama.cpp

The race

No lead changes

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

Beam 1.0 501B A23BSpark X2.6 4B

Every test case

153 test cases

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

Agents101

Structured Extraction161

Coding101

Grounded Answers130

Instruction Following120

Business Logic100

Security & permissions105

Diagnosis & investigation10

82 won by Beam 1.0 501B A23B8 won by Spark X2.6 4B63 tied