Qwen3 Next 80B A3B Thinking in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
C 1,346 ±3
Percentile 46 · #158 · provisional
People prefer / Solves tasks
1,356 / —
Human votes vs benchmarks, Elo-eq
Value
-10
$0.4125 per 1M tokens, median of 13 providers · vendor $1.88
On your hardware
- GPU 12 GB #5 · Q5_K_M · ≈24 tok/s
- GPU 24 GB #11 · UD-Q6_K_XL · ≈26 tok/s
- Mac 64 GB #11 · Q4_K_S · ≈84 tok/s
- 128 GB unified memory #16 · Q8_0 · ≈36 tok/s
- CPU, 32 GB RAM #17 · UD-IQ2_XXS · strong compression · ≈7 tok/s
Score decomposition
C General 1,346 [1,343–1,350] Percentile 46 · #158 · components: 6 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/overall
|
1,368 ±2.99 | 1,368 ±3 | 26.8% | 2026-10-02 | — |
| LMArena |
text/hard_prompts
|
1,371 ±3.85 | 1,369 ±4 | 26.4% | 2026-10-02 | — |
| LMArena |
text/instruction_following
|
1,344 ±5.07 | 1,360 ±5 | 12.6% | 2026-10-02 | — |
| LMArena |
text/expert
|
1,376 ±12 | 1,375 ±10 | 9.8% | 2026-10-02 | — |
| LMArena |
text/longer_query
|
1,354 ±5.61 | 1,358 ±5 | 6.2% | 2026-10-02 | — |
| LMArena |
text/multi_turn
|
1,347 ±6.52 | 1,350 ±6 | 6.0% | 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 Coding 1,356 [1,347–1,365] Percentile 47 · #148 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/coding
|
1,392 ±5.86 | 1,374 ±5 | 89.5% | 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
B Math 1,363 [1,347–1,379] Percentile 50 · #127 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,398 ±10 | 1,392 ±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 1,337 [1,328–1,346] Percentile 46 · #145 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,315 ±7.26 | 1,341 ±7 | 59.8% | 2026-10-02 | — |
| LMArena |
text/longer_query
|
1,354 ±5.61 | 1,358 ±5 | 33.2% | 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: 繁體中文 1,366 [1,353–1,380] Percentile 44 · #125 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,407 ±9.21 | 1,379 ±7 | 92.8% | 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
B Language: English 1,369 [1,361–1,377] Percentile 54 · #127 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,393 ±4.06 | 1,380 ±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
D Language: Español 1,343 [1,320–1,366] Percentile 18 · #123 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/spanish
|
1,357 ±14 | 1,358 ±13 | 90.0% | 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: 한국어 1,350 [1,328–1,373] Percentile 27 · #99 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,309 ±17 | 1,366 ±13 | 90.3% | 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
D Language: Polski 1,348 [1,331–1,365] Percentile 16 · #110 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/polish
|
1,345 ±11 | 1,361 ±10 | 91.9% | 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 Language: Русский 1,340 [1,325–1,356] Percentile 39 · #143 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,339 ±9.37 | 1,352 ±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
History
The same in a table
| Model | 2026-10 |
|---|---|
| Qwen3 Next 80B A3B Thinking | 1,346 C |
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.