Claude Haiku 4.5 in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
B 1,383 ±6
Percentile 60 · #119
People prefer / Solves tasks
1,400 / 1,358
Human votes vs benchmarks, Elo-eq
Value
-12
$2 per 1M tokens, median of 39 providers · vendor $2
On your hardware
Closed weights: cloud only.
Score decomposition
B General 1,383 [1,378–1,389] Percentile 60 · #119 · components: 9
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| Epoch AI |
eci
|
142 ±1.19 | 1,378 ±9 | 31.0% | 2025-10-15 | — |
| LMArena |
text/overall
|
1,396 ±1.24 | 1,396 ±1 | 16.3% | 2026-10-02 | — |
| LMArena |
text/hard_prompts
|
1,420 ±1.55 | 1,412 ±1 | 16.3% | 2026-10-02 | — |
| LMArena |
text/instruction_following
|
1,414 ±1.91 | 1,426 ±2 | 8.1% | 2026-10-02 | — |
| LMArena |
text/expert
|
1,442 ±3.06 | 1,429 ±2 | 8.0% | 2026-10-02 | — |
| Epoch AI |
simpleqa_verified
decay ×0.61 |
0.13 ±0.01 | 1,339 ±3 | 4.2% | 2026-08-10 | — |
| LMArena |
text/longer_query
|
1,427 ±1.91 | 1,426 ±2 | 4.0% | 2026-10-02 | — |
| LMArena |
text/multi_turn
|
1,409 ±2.35 | 1,407 ±2 | 4.0% | 2026-10-02 | — |
| Epoch AI |
gpqa_diamond
decay ×0.26 |
0.71 ±0.03 | 1,371 ±15 | 1.0% | 2025-10-22 | 32k |
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 1,395 [1,393–1,396] Percentile 66 · #95 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
webdev/overall
|
1,329 ±2.49 | 1,401 ±1 | 45.5% | 2026-10-01 | — |
| LMArena |
text/coding
|
1,453 ±2.07 | 1,429 ±2 | 45.0% | 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
B Math 1,367 [1,361–1,373] Percentile 52 · #122 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,395 ±3.65 | 1,389 ±4 | 88.4% | 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
B Writing 1,396 [1,393–1,399] Percentile 63 · #99 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,372 ±2.43 | 1,399 ±2 | 62.4% | 2026-10-02 | — |
| LMArena |
text/longer_query
|
1,427 ±1.91 | 1,426 ±2 | 31.6% | 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 Language: 繁體中文 1,376 [1,371–1,380] Percentile 47 · #119 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,416 ±3.34 | 1,387 ±3 | 94.0% | 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: Deutsch 1,370 [1,359–1,380] Percentile 41 · #86 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/german
|
1,376 ±6.85 | 1,382 ±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
B Language: English 1,390 [1,387–1,394] Percentile 60 · #110 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,413 ±1.70 | 1,402 ±2 | 94.0% | 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: Español 1,403 [1,394–1,412] Percentile 49 · #76 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/spanish
|
1,420 ±5.38 | 1,417 ±5 | 93.5% | 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: Français 1,382 [1,372–1,392] Percentile 34 · #89 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/french
|
1,408 ±5.83 | 1,394 ±5 | 93.4% | 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: 日本語 1,369 [1,357–1,381] Percentile 38 · #80 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/japanese
|
1,339 ±8.60 | 1,382 ±6 | 93.1% | 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: 한국어 1,381 [1,371–1,391] Percentile 38 · #84 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,346 ±7.12 | 1,394 ±5 | 93.4% | 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
D Language: Polski 1,348 [1,340–1,356] Percentile 17 · #109 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/polish
|
1,342 ±5.49 | 1,358 ±5 | 93.6% | 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 Language: Русский 1,378 [1,373–1,382] Percentile 48 · #121 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,380 ±2.71 | 1,389 ±2 | 94.0% | 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
The same in a table
| Model | 2026-10 |
|---|---|
| Claude Haiku 4.5 | 1,383 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.