Models / Qwen3.8-Max (0902)

Alibaba

Qwen3.8-Max (0902)

Qwen3.8-Max (0902) is best for agents and tool use, and reasoning and knowledge; pick it when quality matters more than cost. The cheapest model of similar intelligence is Claude Haiku 5.5, at $0.20 per million tokens against $3.00.

Superseded. Qwen4-Max is Alibaba's newer model in this line, at $2.25 blended: see Qwen4-Max or compare the two.

Results 16 Nov 2026 · 83 results · 8 sources

Price per million tokens
$2.00 in · $6.00 out
Alibaba's list price, 10 Nov 2026 · compare prices50% off in batch
Speed
37 tokens a second; on its test prompt the answer starts after 56.2 s, thinking time included
Artificial Analysis, on its usual API
Intelligence Index
45.4
Artificial Analysis
Spring Prompt overall
21st of 36 score 45 of 100 · how it works
Spring Prompt Score
24 of 100 29th of 32 · 3 of 5 benchmarks · our own tests only
Developer
Alibaba, based in China
Weights not published
Released
3 Aug 2026
1,000,000 tokens of context

Where it stands among the models you could choose

Available models grouped by price, intelligence, speed and developer location. Retired models are excluded.

AmongIntelligence IndexSpring Prompt overallPrice, blendedSpeed
Models available today 15th of 185Claude Opus 5.5 21st of 35Claude Fable 5.2 $3.00 blended, 73rd percentilecheapest: Mistral Nemo 64th of 65Nemotron 3.5 Super VL
Models priced $3–10 per million tokens 9th of 38Claude Opus 5.5 13th of 15Claude Opus 5.5 sets this group 14th of 15Mistral Medium 3.5
Models with similar intelligence sets this group 4th of 9GPT-6 Sol $3.00 blended, 50th percentilecheapest: Claude Haiku 5.5 11th of 11Muse Spark 1.3
Models writing under 50 tokens a second 2nd of 9MiMo-V2.6-Pro – $3.00 blended, 75th percentilecheapest: Phi-4 sets this group
Models from developers based in China 2nd of 75MiMo-V2.6-Pro 5th of 11Kimi K3.1 Pro $3.00 blended, 97th percentilecheapest: Qwen3.5-9B 22nd of 23DeepSeek-V4.1-Flash

The name under each rank is the leader of that group; green is the top quarter of the group and red the bottom quarter, by rank. Price is shown as a percentile: the share of the group that costs less, so lower is cheaper. "Sets this group" marks the measure the group is defined by. Spring Prompt overall is our score out of 100 across every source (how it works). Price is the developer's list price (or the typical OpenRouter host where we haven't read one) for three input tokens to one output; intelligence and speed are from Artificial Analysis (data sourced from Artificial Analysis); "similar intelligence" means within 4 points on its Intelligence Index. Developer locations are where each company is based, not where a model is served. Groups under 3 models are not ranked.

Measured cost per task

API cost recorded in our benchmark runs, excluding judging.

TaskBenchmarkCost per task
Product listingCatalogBench≈ $0.026
DeckDeckBench≈ $0.145

Cheapest model that passes

On our benchmarks that define a pass for each task: what one passing answer cost in our run, failed attempts included, and whether a cheaper model passed.

  • CatalogBench · cost per 1,000 accepted listings
    Qwen3.8-Max (0902) passed 84 of 168 attempts, with no estimate: failed-call costs incomplete. Cheapest: GPT-6 Luna at $0.70 (114 of 168 attempts).
  • DeckBench · cost per presentable deck
    Qwen3.8-Max (0902): $0.44 (2 of 6 tasks), 5 of 12. Cheapest: GPT-6.1 Terra at $0.13 (3 of 6 tasks). It passed at least as often.

Cost per attempt divided by the share that passed, at the run date's prices, excluding judging. Each model at its cheapest setting per passing answer. With a handful of tasks, a pass or two either way is within chance. No estimate where a run had missing outputs, because their cost may not have been recorded.

How good is it, and for what?

Each use case ranks the available models its benchmarks measured, then those priced $3–10 per million tokens. The bar shows its percentile in that field, best to the right. Open a row for the results behind it.

Product listingsTurning a sparse product feed and photos into listings that can go live 26th of 32 available · 14th of 14 at its price
BenchmarkQwen3.8-Max (0902)Rank among availableBest available
CatalogBenchCatalogBench: reliably publish-ready 28.6% 22nd of 32 GPT-6 Astra 75.0%
CatalogBenchCatalogBench (sales brief): reliably publish-ready 0.0% 27th of 32 GPT-6 Astra 69.6%
Decks from an analysisTurning a finished analysis into a deck you could present as it is 21st of 32 available · 10th of 15 at its price
BenchmarkQwen3.8-Max (0902)Rank among availableBest available
DeckBenchDeckBench: deck rating 1,035 21st of 32 Qwen4-Max 1,585
User surveysPlanning a user survey and reading its results without being misled Not measured
  • SurveyBench: has not measured this model.
Marketing planningPlanning a year of ad spend without overspending 27th of 32 available · 13th of 14 at its price
BenchmarkQwen3.8-Max (0902)Rank among availableBest available
ROASBenchROASBench: overall score 25.4 27th of 32 Claude Fable 5.2 61.3
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 10th of 159 available · 7th of 35 at its price
BenchmarkQwen3.8-Max (0902)Rank among availableBest available
Artificial AnalysisArtificial Analysis: Terminal-Bench 4.0 · reported 38.9% 15th of 107 Kimi K3.1 Pro 67.2%
Artificial AnalysisArtificial Analysis: Terminal-Bench 2.1 · reported 88.8% 5th of 107 Claude Fable 5.1 91.4%
Artificial AnalysisArtificial Analysis: τ-bench banking · reported 47.8% 6th of 104 Grok 4.6 50.7%
  • tau2-bench: pending. Its latest results are dated 26 May 2026, before this model was released.
  • Berkeley Function Calling Leaderboard (BFCL) V4: pending. Its latest results are dated 16 Dec 2025, before this model was released.
  • OpenHands Index: pending. Its latest results are dated 30 Jun 2026, before this model was released.
  • Microsoft STATE-Bench: pending. Its latest results are dated 29 May 2026, before this model was released.
  • Vending-Bench 2: has not measured this model.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects 18th of 51 available · 10th of 23 at its price
BenchmarkQwen3.8-Max (0902)Rank among availableBest available
GDP.pdfGDP.pdf: professional document tasks · reported 23.2% 13th of 31 GPT-5.6 Sol 30.7%
  • APEX-Agents: has not measured this model.
  • Remote Labor Index: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge 21st of 187 available · 13th of 39 at its price
BenchmarkQwen3.8-Max (0902)Rank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 45.4 15th of 185 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 43.1% 30th of 186 Claude Opus 5.5 61.4%
Artificial AnalysisArtificial Analysis: GPQA Diamond · reported 92.8% 15th of 168 GPT-6 Astra 96.3%
WritingWhat people prefer in blind comparisons, and judged writing quality 17th of 169 available · 10th of 34 at its price
BenchmarkQwen3.8-Max (0902)Rank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,482 18th of 153 Gemini 4 Argon 1,525
  • UGI Leaderboard: has not measured this model.
Sticking to the factsSummarising without inventing things, and factual answers 36th of 139 available · 10th of 31 at its price
BenchmarkQwen3.8-Max (0902)Rank among availableBest available
Arena (formerly LMArena)Arena Text factuality: overall · reported 1,472 19th of 109 Gemini 4 Argon 1,511
SimpleQA Verified (Epoch AI)SimpleQA Verified: correct answers · reported 47.3% 28th of 68 GPT-6 Astra 75.6%
  • Vectara Hallucination Leaderboard: has not measured this model.
Speed under pressureGood decisions against a real clock (fast chess) Too slow for the clock
  • BulletBench: too slow for Blitz 3+2: lost its first 4 games on time, at a median 7.1 s a move; too slow for Bullet 60s: lost its first 4 games on time, at a median 3.0 s a move; too slow for Lightning 10+1: lost its first 4 games on time, at a median 3.8 s a move.

Nearest alternatives

Against Qwen3.8-Max (0902) at $3.00 blended, Intelligence Index 45.4. "Similar intelligence" means within 4 points or better.

Where to run it

2 hosts on OpenRouter at the standard service tier, with a median of $3.00 blended per million tokens. Cheapest first.

HostQuantisationInputOutputBlended
Alibabanot stated$2.00$6.00$3.00
Alibabanot stated$2.00$6.00$3.00

Prices per million tokens from OpenRouter's public endpoint list, checked 10 Nov 2026.

Against the alternatives

The models you would most likely weigh it against: the leaders of the groups above.

Scroll sideways to see every alternative.

Qwen3.8-Max (0902)Claude Opus 5.5
most intelligent at $3–10 per million tokens
Claude Haiku 5.5
cheapest with similar intelligence
Muse Spark 1.3
fastest with similar intelligence
MiMo-V2.6-Pro
most intelligent from China
Price per million tokens $3.00 $8.00$0.20$2.00$0.54
Tokens a second 37 9624025443
Intelligence Index 45.4 57.643.448.146.3
Spring Prompt overall 45 824059–
Rank among available models, by use case
Product listings 26th of 32 13th20th17th–
Decks from an analysis 21st of 32 7th18th25th–
User surveys – 4th7th––
Marketing planning 27th of 32 9th30th25th–
Agents and tool use 10th of 159 3rd28th13th23rd
Professional work 18th of 51 3rd–8th–
Reasoning and knowledge 21st of 187 1st25th11th13th
Writing 17th of 169 5th–7th23rd
Sticking to the facts 36th of 139 3rd–11th10th
Speed under pressure 36th of 36 –21st33rd–

Shaded figures are better than Qwen3.8-Max (0902)'s. A dash means no result.

Research and examples

In the playground: our pick in 0 of 20 tests, 76% of checks passed →

Where it has been measured

2nd most intelligent of the 75 models from developers based in China, and 2nd most intelligent of the 9 models writing under 50 tokens a second. 9th most intelligent of the 38 models priced $3–10 per million tokens. Among models available today, in the top quarter (by percentile) for agents and tool use, reasoning and knowledge, and writing; in the bottom quarter for product listings, marketing planning and speed under pressure.

Every result

83 published results from 8 sources, each in the source's own units, with the configuration that produced it.

CatalogBench · measured by Spring Prompt · 31 results
MeasureValueRankConfigurationDated
CatalogBench (sales brief): channel rules broken% of products, lower is better 3.0% 32 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): claims to check% of products, lower is better 1.3% 22 of 32 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): content quality% of checks, higher is better 93.6% 33 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): failed outputs% of products, lower is better 3.6% 32 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): missing UK information% of products, lower is better 0.0% 1 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): not findable% of products, lower is better 1.8% 5 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): publish-ready listings% of products, higher is better 7.7% 29 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): reliably publish-ready% of products, higher is better 0.0% 27 of 34 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): unsupported claims% of products, lower is better 84.5% 32 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): unsupported claimsclaims per product, lower is better 4.52 33 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): wrong attributes% of products, lower is better 19.6% 22 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): wrong category or variant% of products, lower is better 2.4% 18 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: channel compliance% of products, higher is better 100.0% 1 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: channel rules broken% of products, lower is better 0.6% 20 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: claims to check% of products, lower is better 0.0% 1 of 32 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: conflicts caught% of conflicts, higher is better 100.0% 1 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: content quality% of checks, higher is better 94.2% 32 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: cost per productUS dollars, lower is better $0.0256 27 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: decision accuracy% of decisions, higher is better 93.3% 30 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: failed outputs% of products, lower is better 3.6% 33 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: field accuracy% of missing fields, higher is better 90.3% 33 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: invented values% of filled values, lower is better 2.7% 26 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: missing UK information% of products, lower is better 0.0% 1 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: not findable% of products, lower is better 3.6% 12 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: publish-ready listings% of products, higher is better 50.0% 22 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: reliably publish-ready% of products, higher is better 28.6% 22 of 34 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: time per productseconds, lower is better 78.4 33 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: unsupported claims% of products, lower is better 26.8% 22 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: unsupported claimsclaims per product, lower is better 0.69 20 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: wrong attributes% of products, lower is better 21.4% 28 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026
CatalogBench: wrong category or variant% of products, lower is better 1.8% 6 of 35 Qwen3.8-Max (0902)
provider default reasoning
30 Sep 2026

Not shown: Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.

DeckBench · measured by Spring Prompt · 17 results
MeasureValueRankConfigurationDated
DeckBench: accurate decks% of tasks, higher is better 50.0% 15 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026
DeckBench: caveat dropped% of tasks, lower is better 16.7% 15 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026
DeckBench: clean layout% of tasks, higher is better 66.7% 8 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026
DeckBench: cost per deckUS dollars, lower is better $0.15 20 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026
DeckBench: deck ratingrating, higher is better 1,035
966–1,109
22 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026
DeckBench: design quality% of the maximum, higher is better 67.0% 16 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026
DeckBench: draft figure quoted% of tasks, lower is better 33.3% 19 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026
DeckBench: findings missing% of tasks, lower is better 0.0% 1 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026
DeckBench: head-to-head win rate% of comparisons, higher is better 60.8% 16 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026
DeckBench: layout defects% of tasks, lower is better 33.3% 8 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026
DeckBench: misleading metric used% of tasks, lower is better 0.0% 1 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026
DeckBench: presentable decks% of tasks, higher is better 33.3% 7 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026
DeckBench: recommendation late or wrong% of tasks, lower is better 0.0% 1 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026
DeckBench: slides needing work% of tasks, lower is better 50.0% 17 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026
DeckBench: time per deckseconds, lower is better 543 33 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026
DeckBench: unsupported claims% of tasks, lower is better 16.7% 15 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026
DeckBench: unsupported numbers% of tasks, lower is better 0.0% 1 of 34 Qwen3.8-Max (0902)
provider default reasoning
6 Oct 2026

Not shown: Not your analysis or your brand: invented companies, six public tasks, one deck per model per task at the provider's default reasoning setting.

ROASBench · measured by Spring Prompt · 9 results
MeasureValueRankConfigurationDated
ROASBench: audience scorescore out of 100, higher is better 39.7 33 of 35 Qwen3.8-Max (0902)
provider default reasoning
29 Sep 2026
ROASBench: business scorescore out of 100, higher is better 15.6 30 of 35 Qwen3.8-Max (0902)
provider default reasoning
29 Sep 2026
ROASBench: consistency scorescore out of 100, higher is better 19.3 25 of 35 Qwen3.8-Max (0902)
provider default reasoning
29 Sep 2026
ROASBench: contribution profitsimulated US dollars, higher is better $99,775 30 of 35 Qwen3.8-Max (0902)
provider default reasoning
29 Sep 2026
ROASBench: cost of a runUS dollars, lower is better $0.73 29 of 35 Qwen3.8-Max (0902)
provider default reasoning
29 Sep 2026
ROASBench: months over budgetmonths of 12, lower is better 2 32 of 35 Qwen3.8-Max (0902)
provider default reasoning
29 Sep 2026
ROASBench: overall scorescore out of 100, higher is better 25.4 30 of 35 Qwen3.8-Max (0902)
provider default reasoning
29 Sep 2026
ROASBench: planning scorescore out of 100, higher is better 54.8 17 of 35 Qwen3.8-Max (0902)
provider default reasoning
29 Sep 2026
ROASBench: return on ad spendprofit per $1 spent, higher is better $0.09 30 of 35 Qwen3.8-Max (0902)
provider default reasoning
29 Sep 2026

Not shown: Not real ad performance: the market is a deterministic simulation of one invented skincare brand.

Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 11 results
MeasureValueRankConfigurationDated
Arena Agent: confirmed task successIPS effect estimate, higher is better 0.05
0.03–0.06
21 of 50 Qwen3.8-Max (0902, arena agent) 2 Oct 2026
Arena Agent: praise over complaintIPS effect estimate, higher is better 0.03
0.01–0.05
22 of 50 Qwen3.8-Max (0902, arena agent) 2 Oct 2026
Arena Agent: steerabilityIPS effect estimate, higher is better 0.03
0.02–0.04
19 of 50 Qwen3.8-Max (0902, arena agent) 2 Oct 2026
Arena Agent: tool groundingIPS effect estimate, higher is better -0.01
-0.01–-0.01
45 of 50 Qwen3.8-Max (0902, arena agent) 2 Oct 2026
Arena Text factuality: overallArena rating, higher is better 1,472
1,467–1,476
27 of 182 Qwen3.8-Max (0902) 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,484
1,475–1,493
21 of 406 Qwen3.8-Max (0902) 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,468
1,458–1,478
20 of 411 Qwen3.8-Max (0902) 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,513
1,501–1,525
24 of 363 Qwen3.8-Max (0902) 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,474
1,466–1,481
31 of 413 Qwen3.8-Max (0902) 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,482
1,476–1,487
23 of 413 Qwen3.8-Max (0902) 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,472
1,464–1,480
21 of 412 Qwen3.8-Max (0902) 2 Oct 2026

Not shown: Not a success rate: the estimated effect of choosing this model as the agent's orchestrator, against an even mix of the models tested, inside Arena's own harness.

Artificial Analysis · reported by Artificial Analysis · 11 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Coding Indexindex score, higher is better 76.2 20 of 187 Qwen3.8-Max (0902) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 45.4 34 of 413 Qwen3.8-Max (0902) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 92.8% 28 of 360 Qwen3.8-Max (0902) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 43.1% 60 of 419 Qwen3.8-Max (0902) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 80.3% 70 of 407 Qwen3.8-Max (0902) 20 Oct 2026
Artificial Analysis: Output speedtokens per second, higher is better 36.8 131 of 137 Qwen3.8-Max (0902) 20 Oct 2026
Artificial Analysis: SciCode% of problems, higher is better 52.1% 88 of 205 Qwen3.8-Max (0902) 20 Oct 2026
Artificial Analysis: Terminal-Bench 2.1% of tasks, higher is better 88.8% 9 of 188 Qwen3.8-Max (0902) 20 Oct 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 38.9% 35 of 204 Qwen3.8-Max (0902) 20 Oct 2026
Artificial Analysis: Time to first answer tokenseconds, lower is better 56.2 122 of 137 Qwen3.8-Max (0902) 20 Oct 2026
Artificial Analysis: τ-bench banking% of tasks, higher is better 47.8% 7 of 175 Qwen3.8-Max (0902) 20 Oct 2026

Not shown: Not your codebase or tools.

GDP.pdf · reported by GDP.pdf · 1 result
MeasureValueRankConfigurationDated
GDP.pdf: professional document tasks% of rubric, higher is better 23.2% 18 of 49 Qwen3.8-Max (0902, xhigh reasoning) 1 Oct 2026

Not shown: Not your documents; 100 tasks, so small differences are noise.

SimpleQA Verified (Epoch AI) · reported by SimpleQA Verified (Epoch AI) · 1 result
MeasureValueRankConfigurationDated
SimpleQA Verified: correct answers% of questions, higher is better 47.3% 31 of 85 Qwen3.8-Max (0902, xhigh reasoning) 2 Sep 2026

Not shown: Not answers grounded in your documents; tests what the model remembers.

tau2-bench · reported by tau2-bench · 2 results
MeasureValueRankConfigurationDated
tau2 v1.0.1, banking knowledge, alltools retrieval: consistency (pass^4)% of tasks, higher is better 35.1% 1 of 21 Qwen3.8-Max (0902, xhigh reasoning, tau2) 3 Aug 2026
tau2 v1.0.1, banking knowledge, alltools retrieval: task success (pass^1)% of tasks, higher is better 55.2% 1 of 21 Qwen3.8-Max (0902, xhigh reasoning, tau2) 3 Aug 2026

Not shown: Not your policies or customers: the tasks and policies are tau2's own, and the customer is simulated by gpt-5.2.

Embed a score badge

A badge for its Spring Prompt Score and each of our benchmarks that ranked Qwen3.8-Max (0902), from the current results. Paste the code into a page or README; it links back to where the result comes from.

Qwen3.8-Max (0902) is #29 of 32 on the Spring Prompt Score (24 of 100, from 3 of 5 of our benchmarks), measured by Spring Prompt Spring Prompt Score · 3 of 5 benchmarks
<a href="https://springprompt.com/score#alibaba-qwen3-8-max-0902"><img src="https://springprompt.com/badge/score/alibaba-qwen3-8-max-0902.svg" alt="Qwen3.8-Max (0902) is #29 of 32 on the Spring Prompt Score (24 of 100, from 3 of 5 of our benchmarks), measured by Spring Prompt" height="20"></a>
Qwen3.8-Max (0902) is #22 of 32 on CatalogBench (Reliably publish-ready, factual brief: 28.6%), measured by Spring Prompt CatalogBench · Reliably publish-ready, factual brief
<a href="https://springprompt.com/benchmarks/catalogbench?focus=alibaba-qwen3-8-max-0902"><img src="https://springprompt.com/badge/catalogbench/alibaba-qwen3-8-max-0902.svg" alt="Qwen3.8-Max (0902) is #22 of 32 on CatalogBench (Reliably publish-ready, factual brief: 28.6%), measured by Spring Prompt" height="20"></a>
Qwen3.8-Max (0902) is #21 of 33 on DeckBench (Deck rating: 1,035), measured by Spring Prompt DeckBench · Deck rating
<a href="https://springprompt.com/benchmarks/deckbench?focus=alibaba-qwen3-8-max-0902"><img src="https://springprompt.com/badge/deckbench/alibaba-qwen3-8-max-0902.svg" alt="Qwen3.8-Max (0902) is #21 of 33 on DeckBench (Deck rating: 1,035), measured by Spring Prompt" height="20"></a>
Qwen3.8-Max (0902) is #28 of 33 on ROASBench (Score: 25.4), measured by Spring Prompt ROASBench · Score
<a href="https://springprompt.com/benchmarks/roasbench?focus=alibaba-qwen3-8-max-0902"><img src="https://springprompt.com/badge/roasbench/alibaba-qwen3-8-max-0902.svg" alt="Qwen3.8-Max (0902) is #28 of 33 on ROASBench (Score: 25.4), measured by Spring Prompt" height="20"></a>

Test your own prompt.

Benchmarks aren't your task. Run your prompt on the models you could use, with every answer checked and costed.