vs
3:12
overall capability wins (0 ties across 15 scored). Llama 4 Maverick leads overall.
- Reasoning0:8
- Coding1:2
- Arena1:0
- Pricing0:3
- Speed0:2
- Specs1:2
Verdict
Llama 4 Maverick leads coding (widest gap: +15.7 on LiveCode); Llama 3.1 405B ranks higher on LMArena (+8 Elo); Llama 4 Maverick offers ~7.8x larger context (1M vs 128K).
- Llama 4 Maverick leads coding (widest gap: +15.7 on LiveCode)
- Llama 3.1 405B ranks higher on LMArena (+8 Elo)
- Llama 4 Maverick offers ~7.8x larger context (1M vs 128K)
- Llama 4 Maverick allows ~8.0x more max output (33K vs 4K)
- Llama 4 Maverick is ~8.4x cheaper on a blended token basis than Llama 3.1 405B
Cost for 1M in + 250K out
Illustrative chat workload at primary-provider list prices.
- Llama 3.1 405B
- $4.38
- Llama 4 Maverick
- $0.48
Llama 4 Maverick is about 9.1x cheaper on this workload.
Price per 1M tokens
Full comparison
Organization
Meta
Meta
Family
Llama / Muse
Llama / Muse
License
Llama 3.1 Community
Llama 4 Community
Open weights
Yes
Yes
Release
Jul 23, 2024
Apr 5, 2025
Knowledge cutoff
-
-
API / provider
Together / Fireworks (ref.)
Together / Fireworks (ref.)
Modalities
text → text
text, image → text
Specs
Context window
128K
1M
Max output
4K
33K
Parameters
405B
400B (17B act.)
Pricing
Input $/1M
$3.50
$0.27
Output $/1M
$3.50
$0.85
Blended $/1M (3∶1)
$3.50
$0.42
Speed
tok/s
40
95
TTFT (s)
0.6
0.35
Reasoning
MMLU-Pro
73.3
80.5
GPQA Diamond
51.5
67.1
Humanity's Last Exam
4.0
4.9
AIME 2025
3.0
19.3
MATH-500
70.3
88.9
Humanity's Last Exam (with tools)
—
—
AA-Omniscience Accuracy
23.2
24.9
AA-LCR v1.1
25.3
50.0
CritPt
—
—
MMMU-Pro
—
62.1
IFBench
39.0
43.0
Chartography
—
—
Chartography (With Tools)
—
—
Coding
SWE-bench Verified
—
21.0
SWE-bench Pro
11.2
5.2
SWE-bench Multilingual
—
—
LiveCodeBench
27.7
43.4
Terminal-Bench 2.1
—
7.9
Aider Polyglot
—
15.6
Terminal-Bench 3
—
—
BigCodeBench
—
49.7
SciCode
29.9
31.7
CursorBench
—
—
SWE-Rebench
—
11.0
NL2Repo-Bench
—
—
DeepSWE
—
—
WebDev Arena
—
—
Terminal-Bench 4.0
—
—
CursorBench 4.0
—
—
FrontierCode v1.1 (Main)
—
—
Arena
LMArena Elo
1,335
1,327
| Metric | Llama 3.1 405B | Llama 4 Maverick | Delta |
|---|---|---|---|
| Identity | |||
| Organization | Meta | Meta | — |
| Family | Llama / Muse | Llama / Muse | — |
| License | Llama 3.1 Community | Llama 4 Community | — |
| Open weights | Yes | Yes | — |
| Release | Jul 23, 2024 | Apr 5, 2025 | — |
| Knowledge cutoff | - | - | — |
| API / provider | Together / Fireworks (ref.) | Together / Fireworks (ref.) | — |
| Modalities | text → text | text, image → text | — |
| Specs | |||
| Context window | 128K | 1M | B +872K |
| Max output | 4K | 33K | B +29K |
| Parameters | 405B | 400B (17B act.) | A +5.00 |
| Pricing | |||
| Input $/1M | $3.50 | $0.27 | B +$3.23 |
| Output $/1M | $3.50 | $0.85 | B +$2.65 |
| Blended $/1M (3∶1) | $3.50 | $0.42 | B +$3.08 |
| Speed | |||
| tok/s | 40 | 95 | B +55 |
| TTFT (s) | 0.6 | 0.35 | B +0.25 |
| Reasoning | |||
| MMLU-Pro | 73.3 | 80.5 | B +7.2 pts |
| GPQA Diamond | 51.5 | 67.1 | B +15.6 pts |
| Humanity's Last Exam | 4.0 | 4.9 | B +0.9 pts |
| AIME 2025 | 3.0 | 19.3 | B +16.3 pts |
| MATH-500 | 70.3 | 88.9 | B +18.6 pts |
| Humanity's Last Exam (with tools) | — | — | — |
| AA-Omniscience Accuracy | 23.2 | 24.9 | B +1.7 pts |
| AA-LCR v1.1 | 25.3 | 50.0 | B +24.7 pts |
| CritPt | — | — | — |
| MMMU-Pro | — | 62.1 | — |
| IFBench | 39.0 | 43.0 | B +4.0 pts |
| Chartography | — | — | — |
| Chartography (With Tools) | — | — | — |
| Coding | |||
| SWE-bench Verified | — | 21.0 | — |
| SWE-bench Pro | 11.2 | 5.2 | A +5.9 pts |
| SWE-bench Multilingual | — | — | — |
| LiveCodeBench | 27.7 | 43.4 | B +15.7 pts |
| Terminal-Bench 2.1 | — | 7.9 | — |
| Aider Polyglot | — | 15.6 | — |
| Terminal-Bench 3 | — | — | — |
| BigCodeBench | — | 49.7 | — |
| SciCode | 29.9 | 31.7 | B +1.8 pts |
| CursorBench | — | — | — |
| SWE-Rebench | — | 11.0 | — |
| NL2Repo-Bench | — | — | — |
| DeepSWE | — | — | — |
| WebDev Arena | — | — | — |
| Terminal-Bench 4.0 | — | — | — |
| CursorBench 4.0 | — | — | — |
| FrontierCode v1.1 (Main) | — | — | — |
| Arena | |||
| LMArena Elo | 1,335 | 1,327 | A +8 Elo |
Llama 3.1 405B
- Context
- 128K / 4K out
- Parameters
- 405B
- Price
- $3.50 / $3.50
- Speed
- 40 tok/s · 0.6s TTFT
- Modalities
- text → text
- License
- Llama 3.1 Community
Largest Llama 3.1 dense instruct model - still used where dense (non-MoE) weights are required.
Llama 4 Maverick
- Context
- 1M / 33K out
- Parameters
- 400B (17B act.)
- Price
- $0.27 / $0.85
- Speed
- 95 tok/s · 0.35s TTFT
- Modalities
- text, image → text
- License
- Llama 4 Community
Meta's open multimodal MoE flagship - 400B total / 17B active with 1M context.
Benchmark charts
Winner bars are emphasized. Per-benchmark deltas sit above each chart.