Navigate
Advertise here

Mercury 2.5 Preview (medium) vs Beam 1.0 501B A23B (medium)

Mercury 2.5 Preview (medium) leads on average score with 6.8 vs 6.6. Mercury 2.5 Preview (medium) has the lower benchmark cost at $0.039 vs $0.499. Beam 1.0 501B A23B (medium) is faster at 6.27s vs 6.66s, with pass rates of 66.7% vs 65.2%.

Last updated at: 2026-10-17

Compared models

Rank
#199
Total Output Tokens
150,107
Response Time (avg)
6.66s
Total Cost
$0.039
Rank
#220
Total Output Tokens
98,521
Response Time (avg)
6.27s
Total Cost
$0.499
Recommended model Mercury 2.5 Preview (medium)

It has the best score here (6.8), while costing about 13.0x less than Beam 1.0 501B A23B (medium).

Detailed comparison

Metric Mercury 2.5 Preview Mercury 2.5 Preview medium Release: 2026-09-02 Beam 1.0 501B A23B Beam 1.0 501B A23B medium Release: 2026-10-17
Score 6.8 6.6
Rank #199 #220
Reliability 10.0 10.0
Consistency 9.6 9.3
Attempts 69/69 69/69
Tests Correct
Attempt pass rate 66.7% 65.2%
Flaky tests 1 2
Total Runs 69 69
Cost per result 0.244 3.561
Total Cost $0.039 $0.499
Input Price $0.040 / 1M $1.000 / 1M
Output Price $0.150 / 1M $3.000 / 1M
Cache Read Price N/A $0.100 / 1M
Cache Write Price N/A N/A
Total Input Tokens 397,291 202,937
Output Tokens 6,355 4,328
Reasoning Tokens 143,752 94,193
Response Time (avg) 6.66s 6.27s
Response Time (max) 74.63s 31.55s
Response Time (total) 153.29s 144.28s
Parameters ~100B 501B total (23B active)
Availability Closed Open source

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

Model generation showcase

Hamster playing table tennis

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

#199 Mercury 2.5 Preview

medium
Cost
$0.001
Time
2.7s
Tokens
2,241 tok

#220 Beam 1.0 501B A23B

medium
Cost
$0.005
Time
11.8s
Tokens
4,267 tok

Top Models by Score

Score vs Total Cost

Response Time (avg)

Score vs Response Time (avg)

Total Output Tokens

Score vs Total Output Tokens

Category Breakdown

Coding Score Consistency Attempt pass rate Flaky tests Tests Correct Response Time (avg) Input Tokens Output Tokens Reasoning Tokens
Mercury 2.5 Preview 7.8 10.0 66.7% 0 3.86s 7,839 552 22,126
Beam 1.0 501B A23B 10.0 10.0 100.0% 0 11.81s 7,791 440 27,204

Quick Compare

Switch Comparison Pair

Mercury 2.5 PreviewmediumvsInklinglowFree AvailableKAT-Coder-Pro V2.5lowvsBeam 1.0 501B A23BmediumBeam 1.0 501B A23BmediumvsStep 3.7 FlashhighMercury 2.5 PreviewmediumvsGPT-6 SolnoneLongCat 2.0lowvsBeam 1.0 501B A23BmediumClaude Opus 4.8nonevsMercury 2.5 PreviewmediumClaude Sonnet 5.5lowvsMercury 2.5 PreviewmediumMercury 2.5 PreviewmediumvsGPT-5.6 LunalowMercury 2.5 PreviewhighvsBeam 1.0 501B A23BmediumEmber-1highvsMercury 2.5 PreviewmediumMercury 2.5 PreviewmediumvsSpace Bunny AlphamaxBeam 1.0 501B A23BmediumvsGLM 5.2none