Compare / head-to-head
GPT-5.4 nanovsLlama 3.3 70B Instruct
GPT-5.4 nano leads 2 of 2 shared benchmarks. GPT-5.4 nano has the larger context window (400K tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
2 – 0
GPT-5.4 nano leads
Cheaper per token
—
list price, input + output
Larger context
GPT-5.4 nano
400K tokens
Provider uptime (30d)
94.3262% · —
OpenAI · Meta
GPT-5.4 nano
OpenAI · released 2026-03-17
textvision
Llama 3.3 70B Instruct
Meta · released 2024-12-06
text
- Context
- 131K
- Max out
- —
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 5 · 4 core
Quality
Benchmark matrix
2 shared · 7 only GPT-5.4 nano · 2 only Llama 3.3 70B Instruct| Benchmark | GPT-5.4 nano | Llama 3.3 70B Instruct | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 2 | ||||
| GPQA Diamond | 82.8% ↗ | 50.5% ↗ | +32.3 pt | GPT-5.4 nano |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 87.8% ↗epoch | 5.1% ↗epoch | +82.7 pt | GPT-5.4 nano |
| Only GPT-5.4 nano reports · 7 | ||||
| FrontierMath Tier 4 v2 (Epoch AI run) | 12.2% ↗epoch | not reported | — | — |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | 44.9% ↗epoch | not reported | — | — |
| OSWorld-Verified | 39% ↗ | not reported | — | — |
| SimpleQA Verified | 11.7% ↗epoch | not reported | — | — |
| SWE-Bench Pro | 52.4% ↗ | not reported | — | — |
| Terminal-Bench 2.0 | 46.3% ↗ | not reported | — | — |
| Toolathlon | 35.5% ↗ | not reported | — | — |
| Only Llama 3.3 70B Instruct reports · 2 | ||||
| MMLU-Pro | not reported | 68.9% ↗ | — | — |
| LMArena Elo | not reported | 1274 ↗ | — | — |
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is GPT-5.4 nano minus Llama 3.3 70B Instruct in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing
Side by side
Official pricing pages and model cards| Spec | GPT-5.4 nano | Llama 3.3 70B Instruct | Edge |
|---|---|---|---|
| Context window | 400K tokens | 131K tokens | GPT-5.4 nano |
| Max output | 128K tokens | — | — |
| Input price / 1M | $0.2 ↗ | — | — |
| Output price / 1M | $1.25 ↗ | — | — |
| Cached input / 1M | $0.02 ↗ | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $1.45 | — | — |
| Modalities | text · vision | text | GPT-5.4 nano |
| Released | 2026-03-17 | 2024-12-06 | — |
| Cited benchmark scores | 10 | 5 | — |
Reliability
Provider status
Last 60 days · refreshed daily on this page · live on /statusMore matchups
GPT-5.4 nano vs …
Models sharing the most benchmarksMore matchups
Llama 3.3 70B Instruct vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run GPT-5.4 nano and Llama 3.3 70B Instruct on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.