Llama 4 Scout (DeepInfra) vs Qwen3.8-2.4T-A95B
Side-by-side comparison of API pricing, specs, benchmarks, and capabilities
| Specification |
by DeepInfra
3 providers:
Groq $0.110/$0.340
Meta $0.110/$0.340
DeepInfra $0.100/$0.300
|
by Alibaba
|
|---|---|---|
| Overview | ||
| Status | Current open weights Open weights | Current open weights Open weights |
| Released | Apr 6, 2025 | Aug 8, 2026 |
| Pricing per million tokens | ||
| Input | $0.100/Mtok | $2.00/Mtok |
| Output | $0.300/Mtok | $6.00/Mtok |
| Blended avg | $0.150/Mtok | $3.00/Mtok |
| Specifications | ||
| Context window | 10M tokens | 1M tokens |
| Parameters | 17B (16 experts) | 2.4T (A95B) |
| Speed (TPS) | — | — |
| Modalities | ||
| Input |
textimage
|
text
|
| Benchmarks sources: Vellum LLM Leaderboard (Jun 2026), Meta model card / Artificial Analysis | ||
| Independent composite scores measured for both models — each on its own scale | ||
| AA Intelligence Index | 10.3 | 57.7 |
| AA Agentic Index | 1.1 | 57.1 |
| AA-Omniscience Index (−100–100) | -52.1 | 4.3 |
| Per-benchmark results published for Llama 4 Scout only — no independent per-benchmark scores exist for Qwen3.8-2.4T-A95B yet | ||
| Avg benchmark score | 46.8 | — |
| Perf / dollar | 312 | — |
| GPQA Diamond | 57.2 | — |
| SWE-Bench Verified | 28 | — |
| LiveCodeBench | 32.8 | — |
| Humanity's Last Exam | 12 | — |
| ARC-AGI 2 | 15 | — |
| AIME 2025 | 45 | — |
| MMMLU | 75 | — |
| BFCL | 58 | — |
| HumanEval | 80 | — |
| MATH 500 | 65 | — |
| Providers | ||
| Available from |
Groq — $0.110/$0.340/Mtok
Meta — $0.110/$0.340/Mtok
DeepInfra — $0.100/$0.300/Mtok
|
Alibaba — $2.00/$6.00/Mtok
|
Cost at scale
1M tokens · 50/50 input/output| Volume | Llama 4 Scout | Qwen3.8-2.4T-A95B | Savings |
|---|---|---|---|
| 1M tokens | $0.15 | $3 | $2.85 (95%) |
| 10M tokens | $1.5 | $30 | $28.5 (95%) |
| 100M tokens | $15 | $300 | $285 (95%) |
| 1000M tokens | $150 | $3000 | $2850 (95%) |
When to pick which
distilled from the pricing and spec data above- cost dominates: $0.150/Mtok blended vs $3.00 — 95% less on the same 50/50 token mix
- you need the longer context: 10M tokens vs 1M (10×)
- provider flexibility: available from 3 providers, so you can price-shop hosts or fail over
- this pairing gives Qwen3.8-2.4T-A95B no edge on price, context, cache discount, or measured performance — pick it only for qualitative fit (ecosystem, compliance, model behavior on your prompts)
Summary
Llama 4 Scout by DeepInfra costs $0.100/Mtok input and $0.300/Mtok output, with a 10M-token context window. It supports text, image input and is available from 3 providers.
Qwen3.8-2.4T-A95B by Alibaba costs $2.00/Mtok input and $6.00/Mtok output, with a 1M-token context window. It supports text input.
On a blended cost basis, Llama 4 Scout is 95% cheaper than Qwen3.8-2.4T-A95B.It also has a larger context window.
The two aren't directly comparable on average benchmark score: Llama 4 Scout has published per-benchmark results, while Qwen3.8-2.4T-A95B does not yet — it is measured today on independent composites (see the table above).
Note: Pricing is per million tokens. Actual costs vary with usage patterns, prompt caching, and batch discounts. Always verify against official provider pricing pages.