Skip to content
fedi.software

Mercury 2 in fedi-index

The score in every slice, what it is made of, value, local run and history.

fedi-index · General

C 1,334 ±6

Percentile 44 · #165 · provisional

People prefer / Solves tasks

1,344 / —

Human votes vs benchmarks, Elo-eq

Value

-20

$0.375 per 1M tokens, median of 3 providers · vendor $0.375

On your hardware

Closed weights: cloud only.

Score decomposition

C General 1,334 [1,328–1,340] Percentile 44 · #165 · components: 5 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,355 ±5.28 1,355 ±5 31.5% 2026-10-02 —
LMArena text/hard_prompts 1,362 ±6.78 1,360 ±6 30.3% 2026-10-02 —
LMArena text/instruction_following 1,328 ±9.76 1,345 ±9 12.8% 2026-10-02 —
LMArena text/longer_query 1,329 ±9.80 1,334 ±9 6.4% 2026-10-02 —
LMArena text/multi_turn 1,358 ±13 1,360 ±12 5.6% 2026-10-02 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

C Coding 1,346 [1,338–1,353] Percentile 42 · #163 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,170 ±12 1,356 ±4 50.2% 2026-10-01 —
LMArena text/coding 1,391 ±10 1,373 ±9 38.7% 2026-10-02 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

C Writing 1,311 [1,296–1,327] Percentile 40 · #162 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,288 ±13 1,314 ±13 54.9% 2026-10-02 —
LMArena text/longer_query 1,329 ±9.80 1,334 ±9 35.8% 2026-10-02 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

C Language: English 1,353 [1,337–1,369] Percentile 46 · #147 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,379 ±8.39 1,366 ±9 92.2% 2026-10-02 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

C Language: Русский 1,308 [1,286–1,331] Percentile 31 · #161 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,303 ±14 1,320 ±12 90.4% 2026-10-02 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

History

fedi-index by month
The same in a table
Model 2026-10
Mercury 2 1,334 C

Badge

Show the model's fedi-index on your site or in a README. The badge updates itself; it links to this page.

fedi-index: Mercury 2

Data CC BY 4.0: keep the link to fedi.software.