Models / GLM-4.6

Z.ai

GLM-4.6

GLM-4.6 is strongest at agents and tool use; worth a look for simple, high-volume work where cost matters most. The cheapest model of similar intelligence is Gemma 4 26B A4B, at $0.14 per million tokens against $0.88.

Superseded. GLM-5.3 is Z.ai's newer model in this line, at $2.15 blended: see GLM-5.3 or compare the two.

Results 16 Nov 2026 · 40 results · 5 sources

Test GLM-4.6 on your prompt → It runs beside the models you could use instead, with every answer checked and costed.

Price per million tokens
$0.50 in · $2.00 out
Typical host on OpenRouter, 10 Nov 2026 · compare prices
Speed
Not measured
Intelligence Index
18.5
Artificial Analysis, at thinking reasoning; 14.9–18.5 across 2 settings
Spring Prompt overall
Not ranked yet
needs our benchmarks and two groups of results
Developer
Z.ai, based in China
Weights published
Released
30 Sep 2025
204,800 tokens of context

Where it stands among the models you could choose

Available models grouped by price, intelligence, speed and developer location, plus models with published weights. Retired models are excluded.

AmongIntelligence IndexSpring Prompt overallPrice, blendedSpeed
Models available today 102nd of 185Claude Opus 5.5 – $0.88 blended, 54th percentilecheapest: Mistral Nemo –
Models priced $0.25–1 per million tokens 26th of 61MiMo-V2.6-Pro – sets this group –
Models with similar intelligence sets this group – $0.88 blended, 62nd percentilecheapest: Gemma 4 26B A4B –
Models from developers based in China 42nd of 75MiMo-V2.6-Pro – $0.88 blended, 66th percentilecheapest: Qwen3.5-9B –
Open-weights models 42nd of 104MiMo-V2.6-Pro – $0.88 blended, 73rd percentilecheapest: Mistral Nemo –

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.

How good is it, and for what?

Each use case ranks the available models its benchmarks measured, then those priced $0.25–1 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 Not measured
  • CatalogBench: has not measured this model.
Decks from an analysisTurning a finished analysis into a deck you could present as it is Not measured
  • DeckBench: has not measured this model.
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 Not measured
  • ROASBench: has not measured this model.
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 61st of 159 available · 8th of 47 at its price
BenchmarkGLM-4.6Rank among availableBest available
Berkeley Function Calling Leaderboard (BFCL) V4BFCL: overall accuracy · reported 72.4% 3rd of 36 Claude Opus 4.5 77.5%
Artificial AnalysisArtificial Analysis: Terminal-Bench 2.1 · reported 49.4% 60th of 107 Claude Fable 5.1 91.4%
Artificial AnalysisArtificial Analysis: τ-bench banking · reported 13.4% 60th of 104 Grok 4.6 50.7%
  • tau2-bench: has not measured this model.
  • OpenHands Index: has not measured this model.
  • Microsoft STATE-Bench: has not measured this model.
  • Vending-Bench 2: has not measured this model.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects Not measured
  • APEX-Agents: has not measured this model.
  • GDP.pdf: has not measured this model.
  • Remote Labor Index: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge 115th of 187 available · 32nd of 61 at its price
BenchmarkGLM-4.6Rank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 18.5 102nd of 185 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 14.5% 116th of 186 Claude Opus 5.5 61.4%
Artificial AnalysisArtificial Analysis: GPQA Diamond · reported 78.0% 102nd of 168 GPT-6 Astra 96.3%
WritingWhat people prefer in blind comparisons, and judged writing quality 77th of 169 available · 13th of 58 at its price
BenchmarkGLM-4.6Rank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,424 77th of 153 Gemini 4 Argon 1,525
UGI LeaderboardUGI: writing score · reported 53.3 52nd of 124 Gemini 3.8 Flash 78.6
Sticking to the factsSummarising without inventing things, and factual answers 68th of 139 available · 12th of 40 at its price
BenchmarkGLM-4.6Rank among availableBest available
Vectara Hallucination LeaderboardVectara: hallucination rate · reported 9.5% 37th of 71 GPT-5.4 nano 3.1%
Arena (formerly LMArena)Arena Text factuality: overall · reported 1,447 54th of 109 Gemini 4 Argon 1,511
  • SimpleQA Verified (Epoch AI): has not measured this model.
Speed under pressureGood decisions against a real clock (fast chess) Not measured
  • BulletBench: has not measured this model.

Nearest alternatives

Against GLM-4.6 at $0.88 blended, Intelligence Index 18.5. "Similar intelligence" means within 4 points or better.

Where to run it

4 hosts on OpenRouter at the standard service tier, with a median of $0.92 blended per million tokens; quantisation where the host states it: FP4, BF16. Cheapest first.

HostQuantisationInputOutputBlended
VeniceFP4$0.43$1.75$0.76
DeepInfraFP4$0.50$2.00$0.88
NovitaBF16$0.55$2.20$0.96
Z.AIFP4$0.60$2.20$1.00

Other service tiers, not counted above: DeepInfra flex $0.40 in, $1.60 out. Flex and batch are cheaper but slower or queued; priority and fast cost more.

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.

GLM-4.6MiMo-V2.6-Pro
most intelligent at $0.25–1 per million tokens
Gemma 4 26B A4B
cheapest with similar intelligence
Claude Opus 5.5
most intelligent available today
Price per million tokens $0.88 $0.54$0.14$8.00
Tokens a second – 438296
Intelligence Index 18.5 46.316.757.6
Spring Prompt overall – ––82
Rank among available models, by use case
Product listings – ––13th
Decks from an analysis – ––7th
User surveys – ––4th
Marketing planning – ––9th
Agents and tool use 61st of 159 23rd105th3rd
Professional work – ––3rd
Reasoning and knowledge 115th of 187 13th110th1st
Writing 77th of 169 23rd82nd5th
Sticking to the facts 68th of 139 10th43rd3rd
Speed under pressure – –––

Shaded figures are better than GLM-4.6's. A dash means no result.

Where it has been measured

26th most intelligent of the 61 models priced $0.25–1 per million tokens. Among models available today, mid-field on every use case measured so far.

Artificial Analysis lists it but has not published a speed for it yet

Every result

40 published results from 5 sources, each in the source's own units, with the configuration that produced it.

Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 7 results
MeasureValueRankConfigurationDated
Arena Text factuality: overallArena rating, higher is better 1,447
1,443–1,451
75 of 182 GLM-4.6 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,412
1,405–1,420
138 of 406 GLM-4.6 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,400
1,392–1,409
109 of 411 GLM-4.6 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,439
1,426–1,451
132 of 363 GLM-4.6 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,412
1,406–1,419
127 of 413 GLM-4.6 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,424
1,420–1,428
123 of 413 GLM-4.6 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,405
1,398–1,412
119 of 412 GLM-4.6 2 Oct 2026

Not shown: Not an error rate: claims that cannot be checked on the web are skipped, and preference still carries most of the weight.

Artificial Analysis · reported by Artificial Analysis · 17 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Coding Indexindex score, higher is better 45.8 108 of 187 GLM-4.6 (thinking reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 14.9 256 of 413 GLM-4.6 (no reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 18.5 223 of 413 GLM-4.6 (thinking reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 63.2% 283 of 360 GLM-4.6 (no reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 78.0% 191 of 360 GLM-4.6 (thinking reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 5.5% 334 of 419 GLM-4.6 (no reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 14.5% 237 of 419 GLM-4.6 (thinking reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 36.7% 227 of 288 GLM-4.6 (no reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 43.4% 180 of 288 GLM-4.6 (thinking reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 26.3% 348 of 407 GLM-4.6 (no reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 54.0% 266 of 407 GLM-4.6 (thinking reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 2.1% of tasks, higher is better 49.4% 109 of 188 GLM-4.6 (thinking reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 28.8% 112 of 282 GLM-4.6 (no reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 25.0% 128 of 282 GLM-4.6 (thinking reasoning) 20 Oct 2026
Artificial Analysis: τ-bench banking% of tasks, higher is better 13.4% 109 of 175 GLM-4.6 (thinking reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 76.9% 108 of 286 GLM-4.6 (no reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 70.5% 127 of 286 GLM-4.6 (thinking reasoning) 20 Oct 2026

Not shown: Not your codebase or tools.

Berkeley Function Calling Leaderboard (BFCL) V4 · reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 8 results
MeasureValueRankConfigurationDated
BFCL: irrelevance detection% correct, higher is better 85.0% 33 of 109 GLM-4.6 (thinking reasoning, bfcl fc) 16 Dec 2025
BFCL: memory% correct, higher is better 55.7% 5 of 109 GLM-4.6 (thinking reasoning, bfcl fc) 16 Dec 2025
BFCL: multi-turn tasks% correct, higher is better 68.0% 5 of 109 GLM-4.6 (thinking reasoning, bfcl fc) 16 Dec 2025
BFCL: overall accuracy% correct, higher is better 72.4% 4 of 109 GLM-4.6 (thinking reasoning, bfcl fc) 16 Dec 2025
BFCL: relevance detection% correct, higher is better 75.0% 62 of 109 GLM-4.6 (thinking reasoning, bfcl fc) 16 Dec 2025
BFCL: single-turn calls (curated)% correct, higher is better 87.6% 25 of 109 GLM-4.6 (thinking reasoning, bfcl fc) 16 Dec 2025
BFCL: single-turn calls (user-contributed)% correct, higher is better 80.9% 8 of 109 GLM-4.6 (thinking reasoning, bfcl fc) 16 Dec 2025
BFCL: web search% correct, higher is better 77.5% 8 of 109 GLM-4.6 (thinking reasoning, bfcl fc) 16 Dec 2025

Not shown: Not general refusal or safety behaviour.

UGI Leaderboard · reported by UGI Leaderboard · 6 results
MeasureValueRankConfigurationDated
UGI: requested-length error% off the requested word count, lower is better 18.0% 158 of 379 GLM-4.6 (no reasoning) 2 Oct 2025
UGI: requested-length error% off the requested word count, lower is better 13.0% 111 of 379 GLM-4.6 (thinking reasoning) 2 Oct 2025
UGI: style adherencescore from 0 to 1, higher is better 0.37 85 of 379 GLM-4.6 (no reasoning) 2 Oct 2025
UGI: style adherencescore from 0 to 1, higher is better 0.38 71 of 379 GLM-4.6 (thinking reasoning) 2 Oct 2025
UGI: writing scorescore out of 100, higher is better 50.0 166 of 379 GLM-4.6 (no reasoning) 2 Oct 2025
UGI: writing scorescore out of 100, higher is better 53.3 149 of 379 GLM-4.6 (thinking reasoning) 2 Oct 2025

Not shown: Not other format limits such as character counts or bullet counts.

Vectara Hallucination Leaderboard · reported by Vectara Hallucination Leaderboard · 2 results
MeasureValueRankConfigurationDated
Vectara: answer rate% of documents, higher is better 94.5% 93 of 108 GLM-4.6 22 Sep 2026
Vectara: hallucination rate% of summaries, lower is better 9.5% 53 of 108 GLM-4.6 22 Sep 2026

Not shown: Not a quality score: a low rate usually means content filters were triggered, and hallucination rates are measured on answered documents only.

Will GLM-4.6 work for your product?

Benchmarks aren't your task. Paste the prompt your product uses and run it beside the models you could use instead, with every answer checked and costed.

New here? Your email starts a free trial. No credit card needed.