Skip to content
fedi.software

Step 2 (16K) in fedi-index

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

fedi-index · General

C 1,295 ±5

Percentile 35 · #192 · provisional

People prefer / Solves tasks

1,302 / —

Human votes vs benchmarks, Elo-eq

Value

-134

$8.02 per 1M tokens, median of 2 providers · vendor $8.02

On your hardware

Closed weights: cloud only.

Score decomposition

C General 1,295 [1,290–1,300] Percentile 35 · #192 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,321 ±4.36 1,321 ±4 28.4% 2026-10-02 —
LMArena text/hard_prompts 1,299 ±7.77 1,304 ±7 25.2% 2026-10-02 —
LMArena text/instruction_following 1,287 ±6.40 1,306 ±6 13.2% 2026-10-02 —
LMArena text/expert 1,278 ±16 1,295 ±13 8.9% 2026-10-02 —
LMArena text/multi_turn 1,292 ±10 1,301 ±9 5.6% 2026-10-02 —
LMArena text/longer_query 1,307 ±11 1,313 ±10 5.3% 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 1,294 [1,278–1,309] Percentile 32 · #189 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,316 ±10 1,307 ±9 87.6% 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 Math 1,287 [1,270–1,303] Percentile 34 · #169 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,304 ±10 1,302 ±10 84.9% 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 1,329 [1,314–1,343] Percentile 44 · #150 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,328 ±11 1,355 ±11 60.1% 2026-10-02 —
LMArena text/longer_query 1,307 ±11 1,313 ±10 31.4% 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

C Language: 繁體中文 1,308 [1,285–1,331] Percentile 29 · #158 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,331 ±16 1,319 ±13 90.1% 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 1,321 [1,310–1,331] Percentile 38 · #169 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,344 ±5.49 1,329 ±6 93.3% 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 1,327 [1,309–1,345] Percentile 33 · #157 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,324 ±11 1,339 ±10 91.7% 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

History

fedi-index by month
The same in a table
Model 2026-10
Step 2 (16K) 1,295 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: Step 2 (16K)

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