Claude Fable 5$22.000/MClaude Opus 4.8$11.000/MClaude Opus 4.7$11.000/MClaude Opus 4.6$11.000/MClaude Opus 4.5$33.000/MClaude Sonnet 3.7$6.600/MClaude Opus 3$33.000/MClaude 2.1$12.800/MClaude 2$12.800/MGPT-5.5$12.500/MGPT-5.2$5.425/MGPT-5.2-Codex$5.425/MGPT-5$3.875/MGPT-4.5$97.500/MGPT-4 Turbo Preview$16.000/MGPT-4$39.000/MGPT-4-32k$78.000/Mo3$19.000/Mo3-mini$2.090/Mo4-mini$2.090/Mo1$28.500/Mo1-mini$5.700/Mo1-preview$28.500/MGemini 3.5 Pro$5.000/MGemini 3.1 Pro$5.000/MGemini 3 Pro$5.000/MGemini 2.5 Pro$3.875/MGemini 1.5 Pro$2.375/MGemini 1.0 Ultra$12.000/MGemini 1.0 Pro$0.800/MClaude Fable 5$22.000/MClaude Opus 4.8$11.000/MClaude Opus 4.7$11.000/MClaude Opus 4.6$11.000/MClaude Opus 4.5$33.000/MClaude Sonnet 3.7$6.600/MClaude Opus 3$33.000/MClaude 2.1$12.800/MClaude 2$12.800/MGPT-5.5$12.500/MGPT-5.2$5.425/MGPT-5.2-Codex$5.425/MGPT-5$3.875/MGPT-4.5$97.500/MGPT-4 Turbo Preview$16.000/MGPT-4$39.000/MGPT-4-32k$78.000/Mo3$19.000/Mo3-mini$2.090/Mo4-mini$2.090/Mo1$28.500/Mo1-mini$5.700/Mo1-preview$28.500/MGemini 3.5 Pro$5.000/MGemini 3.1 Pro$5.000/MGemini 3 Pro$5.000/MGemini 2.5 Pro$3.875/MGemini 1.5 Pro$2.375/MGemini 1.0 Ultra$12.000/MGemini 1.0 Pro$0.800/M
BETA
Subscribe
← Back to Index·All Fireworks AI models
Fireworks AI
Fireworks AIEfficient

Llama 3.1 8B (fw)

Nano

Small, fast Llama 3.1. The default cheap open-weights model in 2024–25 across Groq, Cerebras and Together.

Llama 3.1 8B (fw) is a efficient AI model from Fireworks AI. It costs $0.100 per million input tokens and $0.100 per million output tokens (blended $0.100/M), with a 131K-token context window.

Profile inherited from upstream Llama 3.1 8B ↗ — this is a hosted variant of the same open-weights model.

Released Jul 2024Modalities textOfficial model page ↗Provider pricing ↗API docs ↗Compare with another model →Estimate monthly cost →
INPUT
$0.100/M
per million input tokens
OUTPUT
$0.100/M
per million output tokens
BLENDED 70/30
$0.100/M
default reference rate · how it's calculated →
CONTEXT
131K
131,000 tokens
What it's good at
  • Cheap
  • Fast on LPU/Cerebras
  • Open weights
  • 128K context
Typical use cases
  • Bulk classification
  • Cheap chat
  • Edge inference
Benchmarks
vs. best public score
Scores inherited from Llama 3.1 8B — this is a hosted variant of the same open-weights model, so the underlying benchmark scores are identical.
MMLU73%
Multitask academic knowledge across 57 subjects.
GPQA Diamond31%
Graduate-level science questions, "Google-proof".
MATH51%
High-school competition math problems.
HumanEval73%
Python function synthesis from docstrings.
LMArena Elo1180 Elo
Crowd-sourced head-to-head preference Elo rating.
Hand-curated from each provider's published reports and public leaderboards. Methodology varies across sources — treat as directional rather than authoritative.
More from Fireworks AI
See all 7 →
Llama 3.1 405B (fw)
Serverless · 131K ctx
in $3.000/Mout $3.000/M
Mixtral 8x7B (fw)
Serverless · 32K ctx
in $0.200/Mout $0.200/M
DeepSeek-R1 (fw)
Reasoning · 64K ctx
in $3.000/Mout $8.000/M
Qwen2.5-72B (fw)
Balanced · 32K ctx
in $0.900/Mout $0.900/M
Yi-34B (fw)
Open · 4K ctx
in $0.900/Mout $0.900/M
Phind-CodeLlama-34B (fw)
Code · 16K ctx
in $0.800/Mout $0.800/M
Frequently asked questions

How much does Llama 3.1 8B (fw) cost?

Llama 3.1 8B (fw) costs $0.100 per million input tokens and $0.100 per million output tokens, for a blended reference rate of $0.100 per million tokens.

What is Llama 3.1 8B (fw)'s context window?

Llama 3.1 8B (fw) supports up to 131K tokens of context (131,000 tokens).

What is Llama 3.1 8B (fw) best for?

Llama 3.1 8B (fw) is well suited to Cheap, Fast on LPU/Cerebras and Open weights.

Who makes Llama 3.1 8B (fw)?

Llama 3.1 8B (fw) is developed and served by Fireworks AI. It was released in Jul 2024.