vs
11:5
overall capability wins (1 ties across 17 scored). Claude Opus 5 leads overall.
- Reasoning5:1
- Coding4:0
- Arena1:0
- Agents1:1
- Tool use0:2
- Composite indices2:0
- Pricing0:3
- Speed0:2
- Specs0:1
Verdict
Claude Opus 5 leads coding (widest gap: +17.1 on DeepSWE); Claude Opus 5 ranks higher on LMArena (+12 Elo); Qwen3.8 Max ships open weights (Qwen3.8-Max License).
- Claude Opus 5 leads coding (widest gap: +17.1 on DeepSWE)
- Claude Opus 5 ranks higher on LMArena (+12 Elo)
- Qwen3.8 Max ships open weights (Qwen3.8-Max License)
- Qwen3.8 Max is ~3.3x cheaper on a blended token basis than Claude Opus 5
Cost for 1M in + 250K out
Illustrative chat workload at primary-provider list prices.
- Claude Opus 5
- $11.25
- Qwen3.8 Max
- $3.50
Qwen3.8 Max is about 3.2x cheaper on this workload.
Price per 1M tokens
Full comparison
Organization
Anthropic
Alibaba
Family
Claude Opus
Qwen
License
Proprietary
Qwen3.8-Max License
Open weights
No
Yes
Release
Jul 24, 2026
Aug 3, 2026
Knowledge cutoff
-
-
API / provider
Anthropic
Alibaba Cloud
Modalities
text, image → text
text, image, video → text
Specs
Context window
1M
1M
Max output
128K
131K
Parameters
—
2.4T (95B act.)
Pricing
Input $/1M
$5
$2
Output $/1M
$25
$6
Blended $/1M (3∶1)
$10
$3
Speed
tok/s
72
86
TTFT (s)
0.55
0.4
Reasoning
MMLU-Pro
—
—
GPQA Diamond
93.2
92.8
Humanity's Last Exam
54.9
43.1
AIME 2025
—
—
MATH-500
—
—
Humanity's Last Exam (with tools)
—
—
AA-Omniscience Accuracy
60.9
31.7
AA-LCR v1.1
79.3
80.3
CritPt
29.1
17.7
MMMU-Pro
84.7
82.8
IFBench
—
—
Chartography
—
—
Chartography (With Tools)
—
—
Coding
SWE-bench Verified
—
—
SWE-bench Pro
—
67.7
SWE-bench Multilingual
—
—
LiveCodeBench
—
—
Terminal-Bench 2.1
89.1
81.3
Aider Polyglot
—
—
Terminal-Bench 3
42.7
—
BigCodeBench
—
—
SciCode
56.4
52.1
CursorBench
70.0
—
SWE-Rebench
63.4
—
NL2Repo-Bench
—
55.9
DeepSWE
73.7
56.6
WebDev Arena
1,687
1,671
Terminal-Bench 4.0
51.8
—
CursorBench 4.0
—
—
FrontierCode v1.1 (Main)
—
—
Arena
LMArena Elo
1,493
1,481
| Metric | Claude Opus 5 | Qwen3.8 Max | Delta |
|---|---|---|---|
| Identity | |||
| Organization | Anthropic | Alibaba | — |
| Family | Claude Opus | Qwen | — |
| License | Proprietary | Qwen3.8-Max License | — |
| Open weights | No | Yes | — |
| Release | Jul 24, 2026 | Aug 3, 2026 | — |
| Knowledge cutoff | - | - | — |
| API / provider | Anthropic | Alibaba Cloud | — |
| Modalities | text, image → text | text, image, video → text | — |
| Specs | |||
| Context window | 1M | 1M | tie |
| Max output | 128K | 131K | B +3K |
| Parameters | — | 2.4T (95B act.) | — |
| Pricing | |||
| Input $/1M | $5 | $2 | B +$3 |
| Output $/1M | $25 | $6 | B +$19 |
| Blended $/1M (3∶1) | $10 | $3 | B +$7 |
| Speed | |||
| tok/s | 72 | 86 | B +14 |
| TTFT (s) | 0.55 | 0.4 | B +0.15 |
| Reasoning | |||
| MMLU-Pro | — | — | — |
| GPQA Diamond | 93.2 | 92.8 | A +0.4 pts |
| Humanity's Last Exam | 54.9 | 43.1 | A +11.8 pts |
| AIME 2025 | — | — | — |
| MATH-500 | — | — | — |
| Humanity's Last Exam (with tools) | — | — | — |
| AA-Omniscience Accuracy | 60.9 | 31.7 | A +29.2 pts |
| AA-LCR v1.1 | 79.3 | 80.3 | B +1.0 pts |
| CritPt | 29.1 | 17.7 | A +11.4 pts |
| MMMU-Pro | 84.7 | 82.8 | A +1.9 pts |
| IFBench | — | — | — |
| Chartography | — | — | — |
| Chartography (With Tools) | — | — | — |
| Coding | |||
| SWE-bench Verified | — | — | — |
| SWE-bench Pro | — | 67.7 | — |
| SWE-bench Multilingual | — | — | — |
| LiveCodeBench | — | — | — |
| Terminal-Bench 2.1 | 89.1 | 81.3 | A +7.8 pts |
| Aider Polyglot | — | — | — |
| Terminal-Bench 3 | 42.7 | — | — |
| BigCodeBench | — | — | — |
| SciCode | 56.4 | 52.1 | A +4.3 pts |
| CursorBench | 70.0 | — | — |
| SWE-Rebench | 63.4 | — | — |
| NL2Repo-Bench | — | 55.9 | — |
| DeepSWE | 73.7 | 56.6 | A +17.1 pts |
| WebDev Arena | 1,687 | 1,671 | A +16 Elo |
| Terminal-Bench 4.0 | 51.8 | — | — |
| CursorBench 4.0 | — | — | — |
| FrontierCode v1.1 (Main) | — | — | — |
| Arena | |||
| LMArena Elo | 1,493 | 1,481 | A +12 Elo |
Claude Opus 5
- Context
- 1M / 128K out
- Parameters
- —
- Price
- $5 / $25
- Speed
- 72 tok/s · 0.55s TTFT
- Modalities
- text, image → text
- License
- Proprietary
Anthropic's July 2026 flagship - top Artificial Analysis Intelligence Index and strong agentic coding at Opus 4.8 pricing.
Qwen3.8 Max
- Context
- 1M / 131K out
- Parameters
- 2.4T (95B act.)
- Price
- $2 / $6
- Speed
- 86 tok/s · 0.4s TTFT
- Modalities
- text, image, video → text
- License
- Qwen3.8-Max License
Alibaba's Aug 2026 Max flagship (2.4T MoE / 95B active) — frontier agentic coding via QwenCloud / DashScope with 1M multimodal context and $2/$6 API pricing. Open weights released Aug 12, 2026 as Qwen3.8-2.4T-A95B (text-only, thinking-mandatory, Qwen3.8-Max license); the hosted version adds vision input, non-thinking mode, and built-in tools.
Benchmark charts
Winner bars are emphasized. Per-benchmark deltas sit above each chart.