DeepSeek V3.1 Terminus deepseek-v3.1-terminus
deepseek-ai · deepseek hybrid-reasoning
#45 overall
- params
- 671B (A37B)
- arch
- moe
- context
- 160k
- license
- mit open weights
- released
- Sep 2025
- reasoning
- yes
- downloads/30d
- 12.7k
# architecture
- layers
- 61
- d_model
- 7168
- heads
- 128q / 128kv
- head dim
- 56
- experts
- top-8 of 256 + 1 shared
- vocab
- 129280
- family
- deepseek_v3
signal path of one layer, generated from the registry's structured fields — dimension lines quote the model's real numbers; an MoE trace forks at the router, a dense trace runs straight through. Geometry fields come from the repo's config.json.
# why ranked
Overall: #45 — score 56.7 ● low — a single signal; treat with caution
| signal | weight | input (0–100) |
|---|---|---|
| aa_intelligence | 0.70 | 52.8 |
| bench_composite | 0.30 | 65.7 |
benchmark panel evidence:
| complete panel | score (0–100) |
|---|---|
| aa-coding-v1 | 64.6 |
| aa-math-v1 | 66.8 |
Coding: provisional — score 54.4 ● low — a single signal; treat with caution
provisional: too few independent signals for a numbered position — the score is shown, but this model sorts after every ranked model.
| signal | weight | input (0–100) |
|---|---|---|
| aa_coding | 0.30 | 54.4 |
| aider_polyglot | 0.30 | — |
| swe_bench_verified | 0.40 | — |
missing signals are dropped and the remaining weights renormalized — never imputed.
# trend
methodology changed during this history window; score movement across that boundary is not model movement. See methodology v6.
# benchmarks
| benchmark | score | source | date |
|---|---|---|---|
| AA Coding Index | 43.5 | Artificial Analysis | — |
| AA Intelligence Index | 14.8 | Artificial Analysis | — |
| AA Math Index | 89.7 | Artificial Analysis | — |
# where to run
| provider | quant | ctx | $/M in | $/M out | $/M cache | price src | tps | uptime | ✓ |
|---|---|---|---|---|---|---|---|---|---|
| Featherless | — | — | — | — | via hfrouter | — | — | ||
| DeepInfra | 160k | $0.27 | $0.95 | — | via hfrouter | 69 | — | ||
| SiliconFlow | fp8 | 160k | $0.27 | $1.00 | — | openrouter | — | 99.7% | ✓ |
| StreamLake | unknown | 125k | $0.34 | $1.03 | — | via openrouter | — | 99.8% |
sorted by blended price ((3·input + output) / 4 per 1M) · ✓ = the provider's own catalog confirms the offer · "via …" prices are what the aggregator routing the offer charges, not the provider's own list price
source aliases
- openrouter
deepseek/deepseek-v3.1-terminus