Llama 4 Scout llama-4-scout
meta-llama · llama
#103 overall
- params
- 109B (A17B)
- arch
- moe
- context
- 1280k
- license
- llama4 open weights gated
- released
- Apr 2025
- downloads/30d
- 134.0k
# architecture
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: #103 — score 7.5 ● low — a single signal; treat with caution
| signal | weight | input (0–100) |
|---|---|---|
| aa_intelligence | 0.70 | 7.3 |
| bench_composite | 0.30 | 8.0 |
benchmark panel evidence:
| complete panel | score (0–100) |
|---|---|
| aa-coding-v1 | 16.0 |
| aa-math-v1 | 0.0 |
Coding: provisional — score 5.9 ● 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 | 5.9 |
| 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 | 8.2 | Artificial Analysis | — |
| AA Intelligence Index | 6.5 | Artificial Analysis | — |
| AA Math Index | 14.0 | Artificial Analysis | — |
| Chartqa | 0.9 / 1 | LLM Stats | — |
# where to run
| provider | quant | ctx | $/M in | $/M out | $/M cache | price src | tps | uptime | ✓ |
|---|---|---|---|---|---|---|---|---|---|
| Nscale | 890k | $0.09 | $0.29 | — | via hfrouter | 70 | — | ||
| Amazon Bedrock | — | $0.09 | $0.33 | — | bedrock | — | — | ✓ | |
| DeepInfra | fp8 | 320k | $0.10 | $0.30 | — | deepinfra | — | 80.0% | ✓ |
| DeepInfra | fp8 | 320k | $0.10 | $0.30 | — | deepinfra | — | 99.5% | ✓ |
| DeepInfra | fp8 | 320k | $0.10 | $0.30 | — | deepinfra | — | 95.6% | ✓ |
| Novita | 128k | $0.18 | $0.59 | — | novita | 34 | — | ✓ | |
| Together | 1024k | $0.18 | $0.59 | — | via litellm | — | — | ||
| unknown | 1280k | $0.25 | $0.70 | — | via openrouter | — | — |
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
- aa
llama-4-scout- arena
Llama-4-Scout-17B-16E-Instruct- bedrock
Llama 4 Scout 17B- openrouter
meta-llama/llama-4-scout