DeepSeek V3.1 Terminus in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
B 1,390 ±6
Percentile 62 · #111 · provisional
People prefer / Solves tasks
1,403 / —
Human votes vs benchmarks, Elo-eq
Value
+31
$0.4525 per 1M tokens, median of 8 providers
On your hardware
Does not fit the reference setups.
Score decomposition
B General 1,390 [1,384–1,396] Percentile 62 · #111 · components: 5 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/overall
|
1,420 ±5.16 | 1,420 ±5 | 32.1% | 2026-10-02 | thinking |
| LMArena |
text/hard_prompts
|
1,427 ±7.34 | 1,419 ±7 | 30.1% | 2026-10-02 | thinking |
| LMArena |
text/instruction_following
|
1,406 ±10 | 1,419 ±10 | 12.6% | 2026-10-02 | thinking |
| LMArena |
text/longer_query
|
1,425 ±11 | 1,424 ±11 | 6.0% | 2026-10-02 | thinking |
| LMArena |
text/multi_turn
|
1,411 ±12 | 1,409 ±11 | 5.7% | 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
B Coding 1,377 [1,359–1,395] Percentile 56 · #122 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/coding
|
1,426 ±12 | 1,404 ±11 | 86.5% | 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 Writing 1,407 [1,390–1,424] Percentile 67 · #88 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,404 ±14 | 1,432 ±14 | 55.2% | 2026-10-02 | — |
| LMArena |
text/longer_query
|
1,425 ±11 | 1,424 ±11 | 35.0% | 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: English 1,404 [1,389–1,418] Percentile 64 · #98 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,430 ±7.80 | 1,420 ±8 | 92.5% | 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 |
|---|---|
| DeepSeek V3.1 Terminus | 1,390 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.