Compare / Claude Haiku 5.5 vs GLM-5.3

Claude Haiku 5.5 vs GLM-5.3

Which is cheaper, which is more capable, and which does better at each kind of work, from 21 results on 3 sources that measured both models.

AnthropicClaude Haiku 5.5 Released 7 Oct 2026 · $0.10 in · $0.50 out
Z.aiGLM-5.3 Released 14 Aug 2026 · $1.40 in · $4.40 out
Claude Haiku 5.5 wins 113 too close to callGLM-5.3 wins 7

Of the 21 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.

Results as of 19 November 2026 (catalogue release 2026-11-19-3608133e8331). See also Anthropic vs Z.ai, every model from both labs.

Claude Haiku 5.5 or GLM-5.3? The short answer

They score about the same, so pick Claude Haiku 5.5: it costs 8.1 times less per game of bullet chess on BulletBench.

  • Spring Prompt overall: Claude Haiku 5.5 38, GLM-5.3 35 (out of 100; how the score works).
  • Claude Haiku 5.5 is cheaper to run: $0.002 against $0.016 per game of bullet chess on BulletBench (8.1 times less).
  • GLM-5.3 scores higher on the Artificial Analysis Intelligence Index: 44.8 against 43.4.
  • Claude Haiku 5.5 writes faster: 240 tokens a second against 87.
  • Claude Haiku 5.5 is newer: released 7 Oct 2026, against 14 Aug 2026 for GLM-5.3.
  • On our own benchmarks, on the results both models have: GLM-5.3 does better for marketing planning.
  • On third-party benchmarks, on the results both models have: GLM-5.3 does better for agents and tool use.

Head to head

38
Spring Prompt overall
35
43.4
Intelligence Index
44.8
$0.20
Price per million tokens, blendedlower is better
$2.15
240
Output speed, tokens a second
87
1,000,000
Context window, tokens
1,048,576

Every fact side by side

Claude Haiku 5.5GLM-5.3
Price per million tokens$0.10 in · $0.50 out$1.40 in · $4.40 out
Blended (3 in : 1 out)$0.20$2.15
Cheapest host, blended$0.20 (Amazon Bedrock)$0.40 (Wafer, FP4)
Cost per game of bullet chess (BulletBench)$0.002$0.016
Cached input, per million–$0.26 cached input
Intelligence Index43.444.8
Spring Prompt overall3835
Output speed, tokens a second24087
Context window1,000,000 tokens1,048,576 tokens
Released7 Oct 2026 (newer)14 Aug 2026
WeightsClosedOpen
Developer based inthe USChina

Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.

Which is better for what

Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.

Product listingsTurning a sparse product feed and photos into listings that can go live 23 of 35 not measured
Decks from an analysisTurning a finished analysis into a deck you could present as it is 19 of 35 not measured
User surveysPlanning a user survey and reading its results without being misled 9 of 26 not measured
Marketing planningPlanning a year of ad spend without overspending 33 of 35 32 of 35
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 28 of 159 15 of 159
Professional workReal tasks from banking, consulting and law, business documents and freelance projects not measured 15 of 51
Reasoning and knowledgeHard questions across science, maths and general knowledge 25 of 187 31 of 187
WritingWhat people prefer in blind comparisons, and judged writing quality not measured 24 of 169
Sticking to the factsSummarising without inventing things, and factual answers not measured 51 of 139
Speed under pressureGood decisions against a real clock (fast chess) 24 of 38 36 of 38

From our research

Every shared result

Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.

MeasureClaude Haiku 5.5GLM-5.3
BulletBench measured by Spring Prompt · 3 results
Cost per game · Bullet 60sUS dollars, lower is better 23 · 44 of 62 measured $0.0019no reasoning $0.0156
Ladder Elo · Bullet 60sladder Elo, higher is better 50 · 59 of 62 measured 310no reasoning 33
Games lost on time · Bullet 60s% of games, lower is better 36 · 56 of 62 measured 37.5%no reasoning 83.3%low reasoning
ROASBench measured by Spring Prompt · 3 results
Overall scorescore out of 100, higher is better 34 · 33 of 36 measured 24.1 24.8
Audience scorescore out of 100, higher is better 30 · 29 of 36 measured 48.4 51.0
Business scorescore out of 100, higher is better 32 · 33 of 36 measured 13.5 12.9
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 20 · 17 of 249 measured 43.4max reasoning 44.8max reasoning
Humanity's Last Exam% of questions, higher is better 25 · 38 of 250 measured 44.4%max reasoning 42.3%max reasoning
Terminal-Bench 4.0% of tasks, higher is better 20 · 13 of 120 measured 32.8%max reasoning 41.9%max reasoning
Show all 21 results (12 more)
MeasureClaude Haiku 5.5GLM-5.3
BulletBench measured by Spring Prompt · 2 results
Invalid moves · Bullet 60s% of moves, lower is better 1 · 1 of 62 measured 0.0%no reasoning 0.0%
Median move time · Bullet 60smilliseconds, lower is better 43 · 42 of 62 measured 1.7 sno reasoning 1.6 slow reasoning
ROASBench measured by Spring Prompt · 6 results
Consistency scorescore out of 100, higher is better 33 · 32 of 36 measured 10.6 14.8
Contribution profitsimulated US dollars, higher is better 32 · 34 of 36 measured $32,146 $9,718
Cost of a runUS dollars, lower is better 6 · 12 of 36 measured $0.0313 $0.13
Months over budgetmonths of 12, lower is better 1 · 1 of 36 measured 0 0
Planning scorescore out of 100, higher is better 21 · 36 of 36 measured 54.8 51.7
Return on ad spendprofit per $1 spent, higher is better 32 · 34 of 36 measured $0.03 $0.01
Artificial Analysis reported by Artificial Analysis · 4 results
Long-context reasoning (AA-LCR)% of questions, higher is better 21 · 47 of 241 measured 82.7%max reasoning 79.7%max reasoning
Output speedtokens per second, higher is better 10 · 49 of 78 measured 240max reasoning 87.5low reasoning
SciCode% of problems, higher is better 28 · 12 of 121 measured 55.0%max reasoning 59.0%max reasoning
Time to first answer tokenseconds, lower is better 40 · 62 of 78 measured 6.41low reasoning 25.4low reasoning

Quick answers

Is Claude Haiku 5.5 better than GLM-5.3?

They score about the same, so pick Claude Haiku 5.5: it costs 8.1 times less per game of bullet chess on BulletBench.

Which is cheaper, Claude Haiku 5.5 or GLM-5.3?

Claude Haiku 5.5 is cheaper to run: $0.002 against $0.016 per game of bullet chess on BulletBench (8.1 times less).

Which is faster, Claude Haiku 5.5 or GLM-5.3?

Claude Haiku 5.5 writes about 240 tokens a second on its usual API, against 87 for GLM-5.3 (Artificial Analysis).

Which is better for marketing planning, Claude Haiku 5.5 or GLM-5.3?

GLM-5.3 ranks 32 of 35 models on sale for planning a year of ad spend without overspending, against 33 for Claude Haiku 5.5.

Which is better for agents and tool use, Claude Haiku 5.5 or GLM-5.3?

GLM-5.3 ranks 15 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 28 for Claude Haiku 5.5.

Which is better for reasoning and knowledge, Claude Haiku 5.5 or GLM-5.3?

Claude Haiku 5.5 ranks 25 of 187 models on sale for hard questions across science, maths and general knowledge, against 31 for GLM-5.3.

Which is better for speed under pressure, Claude Haiku 5.5 or GLM-5.3?

Claude Haiku 5.5 ranks 24 of 38 models on sale for good decisions against a real clock (fast chess), against 36 for GLM-5.3.

Which has the bigger context window, Claude Haiku 5.5 or GLM-5.3?

GLM-5.3: 1,048,576 tokens, against 1,000,000 for Claude Haiku 5.5.

Run Claude Haiku 5.5 and GLM-5.3 on your own prompt.

Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.