vs
2:10
overall capability wins (0 ties across 12 scored). Qwen3.8 Flash Next leads overall.
- Reasoning1:5
- Coding0:3
- Tool use0:1
- Pricing0:3
- Speed1:1
- Specs1:1
Verdict
Qwen3.8 Flash Next leads coding (widest gap: +174.8 on WebDev Elo); Qwen3.6 Plus offers ~3.8x larger context (1M vs 262K); Qwen3.8 Flash Next allows ~2.0x more max output (131K vs 66K).
- Qwen3.8 Flash Next leads coding (widest gap: +174.8 on WebDev Elo)
- Qwen3.6 Plus offers ~3.8x larger context (1M vs 262K)
- Qwen3.8 Flash Next allows ~2.0x more max output (131K vs 66K)
- Qwen3.8 Flash Next ships open weights (Qwen Community License 1.0)
- Qwen3.8 Flash Next is ~4.9x cheaper on a blended token basis than Qwen3.6 Plus
Cost for 1M in + 250K out
Illustrative chat workload at primary-provider list prices.
- Qwen3.6 Plus
- $1.25
- Qwen3.8 Flash Next
- $0.27
Qwen3.8 Flash Next is about 4.7x cheaper on this workload.
Price per 1M tokens
Full comparison
Organization
Alibaba
Alibaba
Family
Qwen
Qwen
License
Proprietary
Qwen Community License 1.0
Open weights
No
Yes
Release
Apr 2, 2026
Aug 26, 2026
Knowledge cutoff
-
-
API / provider
Alibaba Cloud
Alibaba (QwenCloud)
Modalities
text, image, video → text
text, image, video → text
Specs
Context window
1M
262K
Max output
66K
131K
Parameters
—
—
Pricing
Input $/1M
$0.50
$0.15
Output $/1M
$3
$0.47
Blended $/1M (3∶1)
$1.13
$0.23
Speed
tok/s
70
77.3
TTFT (s)
0.45
2.79
Reasoning
MMLU-Pro
—
—
GPQA Diamond
88.2
92.3
Humanity's Last Exam
27.8
38.0
AIME 2025
—
—
MATH-500
—
—
Humanity's Last Exam (with tools)
—
—
AA-Omniscience Accuracy
26.4
24.5
AA-LCR v1.1
78.3
79.7
CritPt
2.9
11.1
MMMU-Pro
78.0
79.8
IFBench
75.2
—
Chartography
—
—
Chartography (With Tools)
—
—
Coding
SWE-bench Verified
—
—
SWE-bench Pro
—
—
SWE-bench Multilingual
—
—
LiveCodeBench
—
—
Terminal-Bench 2.1
61.4
86.1
Aider Polyglot
—
—
Terminal-Bench 3
—
—
BigCodeBench
—
—
SciCode
40.7
50.6
CursorBench
—
—
SWE-Rebench
—
—
NL2Repo-Bench
—
—
DeepSWE
—
—
WebDev Arena
1,461
1,635
Terminal-Bench 4.0
—
—
CursorBench 4.0
—
—
FrontierCode v1.1 (Main)
—
—
Arena
LMArena Elo
1,443
—
| Metric | Qwen3.6 Plus | Qwen3.8 Flash Next | Delta |
|---|---|---|---|
| Identity | |||
| Organization | Alibaba | Alibaba | — |
| Family | Qwen | Qwen | — |
| License | Proprietary | Qwen Community License 1.0 | — |
| Open weights | No | Yes | — |
| Release | Apr 2, 2026 | Aug 26, 2026 | — |
| Knowledge cutoff | - | - | — |
| API / provider | Alibaba Cloud | Alibaba (QwenCloud) | — |
| Modalities | text, image, video → text | text, image, video → text | — |
| Specs | |||
| Context window | 1M | 262K | A +738K |
| Max output | 66K | 131K | B +66K |
| Parameters | — | — | — |
| Pricing | |||
| Input $/1M | $0.50 | $0.15 | B +$0.35 |
| Output $/1M | $3 | $0.47 | B +$2.53 |
| Blended $/1M (3∶1) | $1.13 | $0.23 | B +$0.90 |
| Speed | |||
| tok/s | 70 | 77.3 | B +7.30 |
| TTFT (s) | 0.45 | 2.79 | A +2.34 |
| Reasoning | |||
| MMLU-Pro | — | — | — |
| GPQA Diamond | 88.2 | 92.3 | B +4.1 pts |
| Humanity's Last Exam | 27.8 | 38.0 | B +10.2 pts |
| AIME 2025 | — | — | — |
| MATH-500 | — | — | — |
| Humanity's Last Exam (with tools) | — | — | — |
| AA-Omniscience Accuracy | 26.4 | 24.5 | A +1.9 pts |
| AA-LCR v1.1 | 78.3 | 79.7 | B +1.4 pts |
| CritPt | 2.9 | 11.1 | B +8.2 pts |
| MMMU-Pro | 78.0 | 79.8 | B +1.8 pts |
| IFBench | 75.2 | — | — |
| Chartography | — | — | — |
| Chartography (With Tools) | — | — | — |
| Coding | |||
| SWE-bench Verified | — | — | — |
| SWE-bench Pro | — | — | — |
| SWE-bench Multilingual | — | — | — |
| LiveCodeBench | — | — | — |
| Terminal-Bench 2.1 | 61.4 | 86.1 | B +24.7 pts |
| Aider Polyglot | — | — | — |
| Terminal-Bench 3 | — | — | — |
| BigCodeBench | — | — | — |
| SciCode | 40.7 | 50.6 | B +9.9 pts |
| CursorBench | — | — | — |
| SWE-Rebench | — | — | — |
| NL2Repo-Bench | — | — | — |
| DeepSWE | — | — | — |
| WebDev Arena | 1,461 | 1,635 | B +175 Elo |
| Terminal-Bench 4.0 | — | — | — |
| CursorBench 4.0 | — | — | — |
| FrontierCode v1.1 (Main) | — | — | — |
| Arena | |||
| LMArena Elo | 1,443 | — | — |
Qwen3.6 Plus
- Context
- 1M / 66K out
- Parameters
- —
- Price
- $0.50 / $3
- Speed
- 70 tok/s · 0.45s TTFT
- Modalities
- text, image, video → text
- License
- Proprietary
Alibaba hosted flagship (Apr 2026) for agentic coding and multimodality — 1M context via DashScope; always-on reasoning.
Qwen3.8 Flash Next
- Context
- 262K / 131K out
- Parameters
- —
- Price
- $0.15 / $0.47
- Speed
- 77.3 tok/s · 2.79s TTFT
- Modalities
- text, image, video → text
- License
- Qwen Community License 1.0
Open-weight sparse-MoE Flash tier (6B active) previewing the Qwen4 architecture, released Aug 26, 2026 — 262K native context extensible to ~1M with YaRN, tuned for cost efficiency at $0.15/$0.47 per 1M (cache-hit $0.016). Reported total parameter count varies by source (~125-180B). Benchmark figures are Alibaba's self-reported launch numbers (SWE-bench Pro 62.5, HLE 35.9); Artificial Analysis places it at 56 on its Intelligence Index.
Benchmark charts
Winner bars are emphasized. Per-benchmark deltas sit above each chart.