Kimi K3 vs GLM-5.2: Benchmarks, Pricing & Which Wins

Two weeks before Kimi K3 (July 16, 2026) reset expectations for open-weight AI, GLM-5.2 (June 13, 2026) held the crown: Zhipu AI’s 744B coding-first flagship was the highest-scoring open-weight model ever measured. Now the two Beijing labs go head-to-head — and this is the most important comparison in open-weight AI right now, because both models target the same buyer: teams building agents and coding pipelines who refuse to pay closed-model prices.

The independent verdict is already in, and it splits cleanly: K3 is the stronger model. GLM-5.2 is the more practical one. K3 wins every shared capability benchmark and adds native vision. GLM-5.2 costs a third as much, runs nearly three times faster, and — unlike K3 — its MIT-licensed weights are downloadable today. Here’s every number, plus a clear verdict for each type of team.

The two models at a glance

	Kimi K3	GLM-5.2
Maker	Moonshot AI	Zhipu AI (Z.ai)
Released	July 16, 2026	June 13, 2026
Parameters	2.8T MoE (16/896 experts)	744B MoE (~40B active; 753B per Artificial Analysis)
Context window	1M tokens (1,048,576)	1M tokens (1,048,576)
Max output	Not publicly capped	131K tokens
Modalities	Text + image + video	Text only
Reasoning	Always on; low / high / max	High / Max toggle (reasoning_effort)
Input price (per 1M)	$3.00 ($0.30 cached)	$1.40 ($0.26 cached)
Output price (per 1M)	$15.00	$4.40
License	Modified MIT (July 27)	MIT — weights live now
Self-hosting	64+ accelerators recommended	Single 8×H200 node (FP8)
Output speed	~62 tok/s	~168 tok/s
AA Intelligence Index	57 (#4 overall)	51 (top open-weight pre-K3)

Pricing: GLM-5.2 is roughly 3× cheaper per task

Billing tier (per 1M tokens)	Kimi K3	GLM-5.2	GLM cheaper by
Input (standard)	$3.00	$1.40	2.1×
Input (cache hit/read)	$0.30	$0.26	1.2×
Output	$15.00	$4.40	3.4×

On Artificial Analysis’s blended basis (7:2:1 cached:input:output), that’s $2.31 per million tokens for K3 vs $0.90 for GLM-5.2. The number that matters most — cost per completed task — lands at ~$0.94 vs ~$0.32. One dollar buys roughly 67K output tokens from K3 versus 227K from GLM-5.2. For high-volume agent loops, that 3× gap compounds into a serious budget difference.

K3’s counter: Moonshot’s Mooncake inference stack reports >90% cache-hit rates on coding workloads, pulling most real input bills to the $0.30 tier — and K3 needs fewer retries on hard tasks, which partially closes the per-task gap. GLM-5.2’s counter-counter: its High/Max reasoning toggle cuts token burn 2–2.5× on routine work, while K3’s low/high effort modes only shipped post-launch — though K3’s new k3-256k variant (~half quota for ≤256K workloads) now narrows the cost-control gap. Zhipu also sells GLM Coding Plan subscriptions (from roughly $10–18/month) for predictable-cost coding access, and third-party providers list GLM-5.2 as low as $0.77/$2.42 per million tokens.

Benchmarks: K3 wins every shared test

Moonshot’s launch suite ran both models through matched agentic harnesses — this is the cleanest capability comparison available:

Benchmark	Kimi K3	GLM-5.2	K3 margin
DeepSWE	67.5	46.2	+21.3
Program Bench	77.8	63.7	+14.1
Terminal-Bench 2.1	88.3	82.7	+5.6
FrontierSWE	81.2	67.3	+13.9
SWE Marathon	42.0	13.0	+29.0
Automation Bench	30.8	12.9	+17.9
GPQA-Diamond	93.5	91.2	+2.3

The shape of these numbers matters. On bounded coding tasks (Terminal-Bench, Program Bench) GLM-5.2 stays within striking distance. On long-horizon autonomous work — SWE Marathon, Automation Bench — the gap explodes to 3×+. That’s exactly the workload class K3 was built for, and it’s where GLM-5.2’s smaller active-parameter budget shows.

What independent testing says

Artificial Analysis ran both on its standardized Intelligence Index v4.1 (same harness, max effort):

Evaluation	Kimi K3	GLM-5.2 (Max)
Intelligence Index (overall)	57	51
Coding Index	76	69
Agentic Index	50	43
GDPval-AA v2 (Elo)	1,668	1,514
AutomationBench-AA	53%	28%
Terminal-Bench v2.1	85%	78%
GPQA Diamond	94%	89%
Humanity’s Last Exam	44%	40%

To GLM-5.2’s credit: its score of 51 was the highest open-weight result ever before K3 shipped five weeks later. And there’s one metric where GLM-5.2 clearly wins — AA-Omniscience non-hallucination rate: 72% vs K3’s 49%. GLM-5.2 is far more likely to refuse or defer rather than fabricate an answer. If your application punishes confident wrong answers harder than unanswered questions, that 23-point gap deserves real weight.

Zhipu’s own published numbers (not in Moonshot’s suite) are also respectable: SWE-bench Pro 62.1 (ahead of GPT-5.5’s 58.6), AIME 2026 99.2%, and a real-world demo completing root-cause analysis across 740,000 server log lines in one session.

Speed: GLM-5.2 is nearly 3× faster

Artificial Analysis measures GLM-5.2 at roughly 168 output tokens/sec vs K3’s ~62. That’s the largest practical difference between these two models after price. K3 is a deliberate, always-max-thinking engine built for long autonomous runs; GLM-5.2 at High effort is built for interactive loops where a developer is waiting. For chat-facing products and IDE autocomplete-adjacent flows, GLM-5.2 feels dramatically snappier.

Context and output

Both models advertise a 1M-token context window — a genuine tie, and a shared engineering flex (GLM-5.2 uses IndexShare attention to cut per-token FLOPs 2.9× at 1M; K3 uses Kimi Delta Attention for 6.3× faster long-context decoding and scored 90.4 on a full-1M eval with no context management). GLM-5.2 guarantees a 131K max output; K3’s output isn’t publicly capped. Filling the window costs $1.40 with GLM-5.2 versus $3.00 with K3.

Open weights and self-hosting: different leagues of “open”

This is where the comparison stops being about benchmarks:

• GLM-5.2 is open today. Full weights on Hugging Face and ModelScope under a plain MIT license, no regional restrictions. Community quantizations already span BF16 (~1.5TB), FP8 (~750–860GB — fits a single 8×H200 node), and 4-bit GGUF (~376GB, runs on 4×H100 via llama.cpp). Day-one support across vLLM, SGLang, Transformers, and KTransformers, plus an Anthropic-compatible endpoint so Claude Code, Cline, and OpenCode work out of the box.
• Kimi K3 becomes open on July 27, 2026 — 2.8T weights under a Modified MIT license (an attribution clause that triggers only above 100M monthly active users; irrelevant for nearly everyone). But Moonshot recommends supernode deployments of 64+ accelerators. Even MXFP4-quantized, K3’s footprint is ~1.4TB before KV cache.

The translation: a single well-equipped team can self-host GLM-5.2 this afternoon. Self-hosting K3 is a decision for companies and labs with serious infrastructure. If “downloadable, single-node, fine-tunable” is your requirement, GLM-5.2 is the only answer in this comparison.

Multimodality: K3’s clean sweep

GLM-5.2 is text-only. K3 natively understands images and video — which powers use cases GLM-5.2 simply can’t touch: screenshot-to-code, UI mockup implementation, video analysis, and visual debugging. It’s a big part of why K3 holds the #1 Frontend Code Arena ranking while GLM-5.2 doesn’t appear in that arena at all. If your agents look at screens, this section is the whole comparison.

Kimi K3 vs GLM-5.2: which should you choose?

Your situation	Pick
Long-horizon autonomous agent runs (SWE Marathon +29 pts)	Kimi K3
Frontend / UI development (#1 Arena, native vision)	Kimi K3
Hardest reasoning and knowledge work (57 vs 51 Index)	Kimi K3
Multimodal inputs (screenshots, mockups, video)	Kimi K3
Bounded coding & terminal tasks at scale	GLM-5.2 (near-tied, 3× cheaper)
High-volume production workloads ($0.32 vs $0.94/task)	GLM-5.2
Latency-sensitive / interactive products (168 vs 62 tok/s)	GLM-5.2
Self-hosting on a single node, fine-tuning	GLM-5.2 (MIT, today)
Hallucination-sensitive applications (72% vs 49%)	GLM-5.2
Air-gapped deployment needed this week	GLM-5.2 (K3 weights land July 27)
Best absolute capability, cost secondary	Kimi K3

The pattern many teams have converged on: run GLM-5.2 as the high-volume default, escalate to K3 for the hardest, multimodal, or longest-horizon tasks. You capture GLM-5.2’s speed and price on 80–90% of traffic and K3’s ceiling where it matters. And since K3 is free to try in the Kimi app — no credit card, bonus credits via invite link — you can benchmark the escalation tier on your real tasks today. Here’s how to get K3 on your PC.

FAQ

Is Kimi K3 better than GLM-5.2? On capability, yes: 57 vs 51 on the Artificial Analysis Intelligence Index, and K3 wins every shared coding benchmark (DeepSWE +21, SWE Marathon +29, Terminal-Bench +5.6). K3 also adds native vision and video. But GLM-5.2 is ~3× cheaper per task, ~2.7× faster, hallucinates less, and its MIT weights are downloadable today — “better” depends on your workload.

Is GLM-5.2 cheaper than Kimi K3? Yes, significantly: $1.40/$4.40 vs $3.00/$15.00 per million input/output tokens, $0.26 vs $0.30 cached. Blended: ~$0.90 vs ~$2.31 per million. Cost per completed task: ~$0.32 vs ~$0.94 (Artificial Analysis) — roughly 3× cheaper.

Which is faster, Kimi K3 or GLM-5.2? GLM-5.2, by a wide margin: ~168 vs ~62 output tokens per second (Artificial Analysis). K3 defaults to max thinking (low/high modes and a half-quota k3-256k variant now available); GLM-5.2’s High/Max toggle still offers simpler cost control on routine tasks.

Can I self-host Kimi K3 or GLM-5.2? GLM-5.2 today — MIT-licensed weights on Hugging Face, FP8 fits one 8×H200 node, 4-bit GGUF runs on 4×H100. K3’s 2.8T weights release July 27, 2026 (Modified MIT), but Moonshot recommends 64+ accelerators, putting self-hosting out of reach for most teams.

Kimi K3 vs GLM-5.2 for coding — which wins? K3 on capability (every shared coding benchmark, #1 Frontend Code Arena, vision for screenshot-to-code). GLM-5.2 on economics for bounded tasks — nearly tied on Terminal-Bench 2.1 at a third of the cost and 2.7× the speed. Long-horizon autonomous coding goes to K3 decisively (SWE Marathon 42.0 vs 13.0).

Should I use both together? Yes — a common production pattern: GLM-5.2 as the high-volume default, K3 for the hardest reasoning, multimodal inputs, and long autonomous runs. Both expose OpenAI-compatible endpoints, so routing between them is a config change, not a rewrite.

The verdict

Kimi K3 vs GLM-5.2 isn’t really a fight — it’s a stack. K3 is the strongest open-weight model ever measured: it sweeps every shared benchmark, sees images and video, and on July 27 becomes the largest downloadable model in history. GLM-5.2 is the most practical open-weight model ever shipped: a third of the price, nearly 3× the speed, fewer hallucinations, plain-MIT weights that fit on one node, and a reasoning toggle for cost control — available right now.

If you can only pick one: GLM-5.2 for volume, K3 for ceiling. For the closed-model fights, see our Kimi K3 vs Claude Opus 5 and Kimi K3 vs Claude Fable 5 breakdowns — or start from the complete Kimi K3 guide.