Claude Opus 4.8 in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
A 1,461 ±6
Percentile 94 · #18
People prefer / Solves tasks
1,454 / 1,470
Human votes vs benchmarks, Elo-eq
Value
+27
$10 per 1M tokens, median of 38 providers · vendor $10
On your hardware
Closed weights: cloud only.
Score decomposition
A General 1,461 [1,455–1,467] Percentile 94 · #18 · components: 9
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| Epoch AI |
eci
|
158 ±1.21 | 1,503 ±10 | 31.0% | 2026-05-28 | — |
| LMArena |
text/overall
|
1,461 ±1.98 | 1,461 ±2 | 16.3% | 2026-10-02 | high |
| LMArena |
text/hard_prompts
|
1,482 ±2.33 | 1,469 ±2 | 16.2% | 2026-10-02 | high |
| LMArena |
text/instruction_following
|
1,476 ±2.72 | 1,485 ±3 | 8.0% | 2026-10-02 | high |
| LMArena |
text/expert
|
1,502 ±4.01 | 1,478 ±3 | 7.9% | 2026-10-02 | high |
| LMArena |
text/longer_query
|
1,483 ±2.62 | 1,478 ±2 | 4.0% | 2026-10-02 | high |
| LMArena |
text/multi_turn
|
1,476 ±3.41 | 1,468 ±3 | 4.0% | 2026-10-02 | high |
| Epoch AI |
simpleqa_verified
decay ×0.61 |
0.53 ±0.02 | 1,464 ±5 | 4.0% | 2026-08-27 | max |
| Epoch AI |
gpqa_diamond
decay ×0.26 |
0.91 ±0.02 | 1,463 ±9 | 1.4% | 2026-06-07 | max |
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 Coding 1,439 [1,436–1,441] Percentile 89 · #32 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
webdev/overall
|
1,556 ±2.92 | 1,465 ±1 | 45.8% | 2026-10-01 | high |
| LMArena |
text/coding
|
1,490 ±2.96 | 1,462 ±3 | 44.6% | 2026-10-02 | high |
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 Agents 1,465 [1,459–1,471] Percentile 84 · #10 · components: 2
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
agent/overall
|
0.07 ±0.01 | 1,487 ±3 | 66.0% | 2026-10-02 | high |
| τ²-bench (Sierra) |
tau2/banking_knowledge
decay ×0.76 |
0.40 ±0.05 | 1,480 ±8 | 26.8% | 2026-07-23 | max |
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
S Math 1,462 [1,455–1,469] Percentile 95 · #13 · components: 3
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,486 ±5.74 | 1,476 ±6 | 52.1% | 2026-10-02 | high |
| Epoch AI |
frontiermath_t1_3
decay ×0.86 |
0.80 ±0.02 | 1,494 ±5 | 30.4% | 2026-06-10 | max |
| Epoch AI |
frontiermath_t4
decay ×0.86 |
0.56 ±0.08 | 1,482 ±12 | 10.0% | 2026-06-10 | max |
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 Vision 1,455 [1,451–1,459] Percentile 84 · #19 · components: 3 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
vision/overall
|
1,295 ±3.31 | 1,470 ±3 | 63.1% | 2026-10-02 | high |
| LMArena |
vision/ocr
|
1,303 ±3.38 | 1,474 ±3 | 15.6% | 2026-10-02 | high |
| LMArena |
vision/diagram
|
1,315 ±4.87 | 1,476 ±4 | 15.1% | 2026-10-02 | high |
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 Search 1,431 [1,425–1,438] Percentile 62 · #13 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
search/overall
|
1,204 ±3.27 | 1,446 ±3 | 93.8% | 2026-08-24 | — |
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 1,465 [1,461–1,470] Percentile 92 · #22 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,455 ±3.35 | 1,484 ±3 | 62.1% | 2026-10-02 | high |
| LMArena |
text/longer_query
|
1,483 ±2.62 | 1,478 ±2 | 31.8% | 2026-10-02 | high |
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: 繁體中文 1,444 [1,437–1,451] Percentile 85 · #34 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,508 ±4.94 | 1,460 ±4 | 93.7% | 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
A Language: Deutsch 1,444 [1,429–1,459] Percentile 88 · #18 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/german
|
1,471 ±9.67 | 1,464 ±8 | 92.4% | 2026-10-02 | high |
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 1,445 [1,440–1,450] Percentile 88 · #33 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,468 ±2.55 | 1,460 ±3 | 93.9% | 2026-10-02 | high |
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: Español 1,442 [1,429–1,454] Percentile 85 · #23 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/spanish
|
1,466 ±7.59 | 1,460 ±7 | 92.9% | 2026-10-02 | high |
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: Français 1,443 [1,431–1,456] Percentile 84 · #23 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/french
|
1,482 ±7.39 | 1,462 ±7 | 93.0% | 2026-10-02 | high |
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: 日本語 1,436 [1,421–1,451] Percentile 83 · #23 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/japanese
|
1,438 ±11 | 1,455 ±8 | 92.4% | 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
A Language: 한국어 1,442 [1,428–1,456] Percentile 85 · #21 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,433 ±9.92 | 1,460 ±8 | 92.7% | 2026-10-02 | high |
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: Polski 1,438 [1,424–1,452] Percentile 83 · #23 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/polish
|
1,458 ±9.11 | 1,457 ±8 | 92.7% | 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
A Language: Русский 1,456 [1,450–1,463] Percentile 92 · #20 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,474 ±3.93 | 1,473 ±4 | 93.8% | 2026-10-02 | high |
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 Opus 4.8 | 1,461 A |
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.