CADLE CAP
97
Cadle #6/385 · 84% confidence
Alibaba · Open weightsQwen/Qwen3.8-Max
Active
Model portfolio
Qwen/Qwen3.8-Max
Tap a star to rate
Alibaba · Released 2026-09-02 · v2026-09-02 · Open · Context 1M tokens
Qwen/Qwen3.8-Max is an open-source model from Alibaba (released 2026-09-02). Context window ~1M tokens with capabilities: Text. List pricing ≈ $1.65 / $4.95 per 1M input/output tokens. Top tracked signals: MATH · epoch 99.4, MATH 98.6, MATH · benchlm 97.7. Cadle merges provider listings, unified leaderboard medians, and public pricing where available.
Language Cap
97
Cadle Language #6/385 · 84%
Benchmark average
87.5
#13 on the benchmark matrix
Pricing
$1.65
$4.95 / 1M out · 14.1 pts/$
CROSS-BOARD RANKS
Cadle Language LLM #6/385Matrix #13/443BenchLM #18/224AA #36/698vs BenchLM: -13
CADLE CAP vs MODEL MATRIX
Cadle Cap ranks models on every published score we have for them — harder benchmarks weigh more. Missing tests are skipped, not zeroed. It answers who is stronger than whom in today's pool.
Model Matrix is the datasheet: Overall, Intelligence, Coding, Agentic, MMLU, GPQA, Math, LiveCodeBench, HumanEval, Reason, ARC-AGI. A dash means unmeasured. AVG is an importance-weighted mean of that model's filled cells only.
MODEL MATRIX · TRUTH CENTER CELLS
Overall
69.7
AA IQ
45.4
Coding
84.6
Agentic
87.7
MMLU
88.7
GPQA
92.8
Math
98.6
LiveCodeBench
70.5
HumanEval
—
Reason
94.7
ARC-AGI
—
Coverage AVG
87.5
Cadle Cap ranks models with within-benchmark percentiles so ARC-AGI ~40 and MMLU ~90 never compete raw. Coverage AVG is an importance-weighted mean of filled expert cells only — blanks are skipped, not zeroed.
ANALYSIS · TRUTH CENTER
Hard benches (GPQA · ARC-AGI · LCB · Agentic) sit left — a mid-40 on ARC-AGI can still be frontier. Broader benches on the right are easier and often higher.
CADLE INSIGHTS
Only measured signals
INTELLIGENCE
54.1
Media Cap quality axis
CODING
84.6
Media Cap quality axis
AGENTIC
87.7
Media Cap quality axis
MATH
98.6
Media Cap quality axis
MATRIX OVERALL
69.7
AVG 87.5 across matrix cells
BENCHMARK AVG
83.0
24 tracked scores
STRONGEST SIGNAL
MATH · epoch
Score 99.4
SPEED · COST · QUALITY
Measured matrix / AA / gateway rows only.
- Overall (Matrix)
- 69.7
- Cap Intelligence
- 93.2
- AA / Intel index
- 45.4
- Cap Coding
- 98.8
- Matrix Coding
- 84.6
- Speed tok/s
- 39
- Latency
- 2.0 s
- Input $/1M
- $1.65
- Output $/1M
- $4.95
- Value pts/$
- 14.12
DESCRIPTION + DETAILS
Qwen/Qwen3.8-Max is an open-source model from Alibaba (released 2026-09-02). Context window ~1M tokens with capabilities: Text. List pricing ≈ $1.65 / $4.95 per 1M input/output tokens. Top tracked signals: MATH · epoch 99.4, MATH 98.6, MATH · benchlm 97.7. Cadle merges provider listings, unified leaderboard medians, and public pricing where available.
PROVIDER
Alibaba
RELEASE
2026-09-02
CONTEXT
1M tokens
STATUS
ACTIVE
COMMUNITY REVIEWS
OPERATOR NOTES
Quality, latency, cost, and failure modes — not marketing copy.
No reviews yet
WRITE A REVIEW
Tap a star to rate
Stored on this device · synced when signed in
No operator notes yet
First review sets the tone for how teams evaluate Qwen/Qwen3.8-Max.
MODEL COMPARISON
| Metric | |
|---|---|
| Provider | Alibaba |
| Release | 2026-09-02 |
| Input / 1M | $1.65 |
| Output / 1M | $4.95 |
| Context | 1M tokens |
| Cadle Cap | 97 |
| Cap board | Cadle Language #6/385 |
| MATH · epoch | 99.4 |
| MATH | 98.6 |
| MATH · benchlm | 97.7 |
| REASON · benchlm | 94.7 |
| REASON | 94.7 |
| GPQA · benchlm | 93.7 |
| CODING · benchlm | 93.0 |
| GPQA | 92.8 |
| GPQA · artificial-analysis | 92.8 |
| GPQA · epoch | 92.7 |
| AGENTIC · artificial-analysis | 88.8 |
| MMLU · benchlm | 88.7 |
DATA PROVENANCE
Scores may use different scales — check the Benchmarks tab for source links.
- OpenRouter / provider APIs (pricing & context)
- Hugging Face Hub (downloads & cards)
- Papers With Code (task benchmarks)
- BenchLM.ai (category rollups + atomic benches)
- Epoch AI Benchmarking Hub (ECI + ZIP CSVs)
- LiveCodeBench (generation / execution / repair / test-output)
- Artificial Analysis (intelligence / indexes)
- LLM Stats (leaderboard + rankings)
- Cadle unified median (equal-weight multi-source)
- Artificial Analysis — throughput, latency, and $/M token snapshots
- Epoch AI Benchmarking Hub — capability CSV exports (when synced)
SOURCE TRACKING
Evidence linked to this model — listing, hub, benchmarks, or live gateway prices.
API
Alibaba listing
API
artificial-analysis
API
OpenRouter
COMMUNITY
Papers with Code
COMMUNITY
BenchLM
COMMUNITY
epoch
COMMUNITY
unified-median
MEASURED CELLS
Same Truth Center numbers as the datasheet above. Accuracy, reliability, safety, and cost scores are not invented.
Overall69.7
AA IQ45.4
Coding84.6
Agentic87.7
MMLU88.7
GPQA92.8
Math98.6
LiveCodeBench70.5
AA Index45.4