Kimi K2.5 in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
B 1,418 ±5
Percentile 78 · #66
People prefer / Solves tasks
1,430 / 1,405
Human votes vs benchmarks, Elo-eq
Value
+36
$1.20 per 1M tokens, median of 33 providers
On your hardware
Does not fit the reference setups.
Score decomposition
B General 1,418 [1,414–1,423] Percentile 78 · #66 · components: 8
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| Epoch AI |
eci
|
148 ±0.74 | 1,422 ±6 | 38.6% | 2026-01-27 | — |
| LMArena |
text/overall
|
1,445 ±1.67 | 1,445 ±2 | 14.6% | 2026-10-02 | thinking |
| LMArena |
text/hard_prompts
|
1,453 ±2.02 | 1,443 ±2 | 14.6% | 2026-10-02 | thinking |
| LMArena |
text/instruction_following
|
1,431 ±2.46 | 1,442 ±2 | 7.2% | 2026-10-02 | thinking |
| LMArena |
text/expert
|
1,466 ±3.88 | 1,448 ±3 | 7.1% | 2026-10-02 | thinking |
| Epoch AI |
simpleqa_verified
decay ×0.61 |
0.34 ±0.02 | 1,405 ±5 | 4.1% | 2026-08-27 | — |
| LMArena |
text/longer_query
|
1,445 ±2.41 | 1,442 ±2 | 3.6% | 2026-10-02 | thinking |
| LMArena |
text/multi_turn
|
1,444 ±3.05 | 1,439 ±3 | 3.6% | 2026-10-02 | thinking |
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
B Coding 1,423 [1,418–1,428] Percentile 80 · #57 · components: 3
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
webdev/overall
|
1,436 ±2.65 | 1,431 ±1 | 41.6% | 2026-10-01 | thinking |
| LMArena |
text/coding
|
1,483 ±6.83 | 1,455 ±6 | 35.8% | 2026-10-02 | instant |
| Epoch AI |
swe_bench_verified
decay ×0.37 |
0.74 ±0.02 | 1,452 ±10 | 14.0% | 2026-02-17 | — |
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
A Math 1,430 [1,422–1,437] Percentile 85 · #38 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,470 ±4.77 | 1,460 ±5 | 88.0% | 2026-10-02 | thinking |
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
B Vision 1,432 [1,429–1,436] Percentile 69 · #36 · components: 3 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
vision/overall
|
1,269 ±2.90 | 1,447 ±3 | 62.9% | 2026-10-02 | thinking |
| LMArena |
vision/ocr
|
1,276 ±2.89 | 1,448 ±3 | 15.7% | 2026-10-02 | thinking |
| LMArena |
vision/diagram
|
1,282 ±4.02 | 1,448 ±4 | 15.3% | 2026-10-02 | thinking |
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
A Writing 1,434 [1,430–1,438] Percentile 80 · #53 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,423 ±3.15 | 1,452 ±3 | 62.1% | 2026-10-02 | thinking |
| LMArena |
text/longer_query
|
1,445 ±2.41 | 1,442 ±2 | 31.8% | 2026-10-02 | thinking |
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
B Language: 繁體中文 1,434 [1,427–1,440] Percentile 80 · #46 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,494 ±4.66 | 1,449 ±4 | 93.8% | 2026-10-02 | thinking |
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
B Language: Deutsch 1,420 [1,406–1,434] Percentile 71 · #43 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/german
|
1,441 ±8.96 | 1,438 ±8 | 92.7% | 2026-10-02 | thinking |
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
A Language: English 1,430 [1,426–1,434] Percentile 82 · #50 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,453 ±2.25 | 1,444 ±2 | 94.0% | 2026-10-02 | thinking |
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
B Language: Español 1,428 [1,417–1,439] Percentile 75 · #38 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/spanish
|
1,449 ±6.53 | 1,444 ±6 | 93.2% | 2026-10-02 | thinking |
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
B Language: Français 1,420 [1,409–1,431] Percentile 63 · #51 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/french
|
1,454 ±6.71 | 1,436 ±6 | 93.2% | 2026-10-02 | thinking |
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
B Language: 日本語 1,424 [1,409–1,439] Percentile 76 · #31 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/japanese
|
1,421 ±11 | 1,442 ±8 | 92.4% | 2026-10-02 | thinking |
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
B Language: 한국어 1,427 [1,415–1,440] Percentile 75 · #35 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,412 ±8.81 | 1,444 ±7 | 93.0% | 2026-10-02 | thinking |
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
B Language: Polski 1,430 [1,418–1,441] Percentile 75 · #34 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/polish
|
1,446 ±7.58 | 1,446 ±6 | 93.1% | 2026-10-02 | thinking |
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
B Language: Русский 1,424 [1,419–1,430] Percentile 76 · #56 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,436 ±3.45 | 1,439 ±3 | 93.9% | 2026-10-02 | thinking |
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
The same in a table
| Model | 2026-10 |
|---|---|
| Kimi K2.5 | 1,418 B |
Badge
Show the model's fedi-index on your site or in a README. The badge updates itself; it links to this page.
Data CC BY 4.0: keep the link to fedi.software.