GLM-5.2 vs Kimi K3: Open Weights vs API-Only Frontier (2026)

TL;DR

Last updated July 17, 2026.

GLM-5.2 is Zhipu AI's 753B open-weight (MIT) coding model, released June 13, 2026, scoring 62.1 on SWE-bench Pro and 81.0 on Terminal-Bench 2.1 at about $1.40/M input and $4.40/M output. Kimi K3 is Moonshot's 2.8-trillion-parameter flagship, launched July 16 as API-only (weights due July 27), with vendor-reported 81.2 on FrontierSWE and 88.3 on Terminal-Bench 2.0 at $3/$15 per M tokens, thinking always on, and native vision. K3 is the capability-and-vision pick; GLM-5.2 is the price-and-open-weights pick.

The 30-Second Verdict

These two models sit at opposite ends of nearly every axis, so the choice is rarely close once you know which axis you are on.

• Optimizing for cost per task? GLM-5.2. Roughly one-third the output price on its hosted API, and you can cap reasoning effort at high; K3 is locked to max at launch.
• Optimizing for peak capability? Kimi K3. It posts the stronger agentic numbers, and the independent Intelligence Index (57 vs 51) backs the direction, even if the vendor tables overstate the gap.
• Need vision? Kimi K3 only. GLM-5.2 is text-only.
• Need weights today? GLM-5.2 only. K3 is API-only until its July 27 weights release.
• Long-context-heavy workload? Both run 1M context; K3 prices it flat at $3/$15, GLM-5.2 is cheaper per token but you serve the context budget against a verbose model.

Spec-by-Spec

The parameter counts tell the story in one line: K3 is nearly four times the total size, which is why it costs more to serve and why self-hosting it is a different weight class. Active parameters, the number that drives per-token compute, are closer, roughly 50B for K3 (estimated) versus about 40B for GLM-5.2.

Both hit a 1M-token context, but they get there differently. GLM-5.2 uses IndexShare, a lightweight sparse-attention indexer reused across every four layers, which Z.ai reports cuts per-token FLOPs by 2.9x at 1M length. K3 uses Kimi Delta Attention, a hybrid linear-attention design that interleaves linear and full-attention layers in a 3:1 ratio and cuts KV-cache memory by up to 75%, which is what lets Moonshot price the 1M window flat. Full details on each are on the GLM-5.2 and Kimi K3 pages.

Benchmarks: Vendor-Reported vs Independent

This is the section where care matters most, because the two models publish different benchmarks on different harness versions. GLM-5.2 reports SWE-bench Pro and Terminal-Bench 2.1; K3 reports FrontierSWE and Terminal-Bench 2.0. Terminal-Bench 2.0 and 2.1 are different harnesses, so the 88.3-vs-81.0 gap is not a clean head-to-head. Both vendors self-report, so treat every vendor row as a claim.

The one apples-to-apples number is Terminal-Bench 2.1 run independently: Z.ai reports 81.0 for GLM-5.2, and Vals independently measured 80.9 for K3. On the same harness version, they are effectively tied on terminal-agent tasks, which reframes the vendor-table gap. Where K3 genuinely pulls ahead is the broad Artificial Analysis Intelligence Index, 57 to GLM-5.2's 51, the open-lineage leader. See our SWE-bench Pro breakdown for how these harnesses behave and where scaffolds inflate scores.

The honest read: K3 is the stronger model, but by less than its vendor table suggests. The FrontierSWE and BrowseComp numbers are Moonshot-reported and still awaiting broad independent replication on SWE-bench-style harnesses. Launch-day Hacker News skepticism centered on benchmark contamination in open-lineage models, and early Arena aggregation put K3 around 1486 on text and 1530 on coding, one tier below the closed frontier rather than at parity.

Price and Token Economics

On sticker price, K3 is about 3x GLM-5.2 on output and roughly 2x on input. But both models are heavy reasoners, so the effective cost per task is what matters, and both have a verbosity tax that the sticker hides.

The verbosity picture cuts against both, but K3 has less headroom. Artificial Analysis measured GLM-5.2 at about 43,000 output tokens per Index task at max effort, 37,000 of them pure reasoning, which pushes its effective cost near $0.46 per task despite the cheap per-token rate. The mitigation is real: GLM-5.2 "high" cuts token usage 2 to 2.5x with little quality drop, and effort is an opt-in you control. K3 gives you no such lever at launch. Thinking mode cannot be disabled and reasoning_effort supports only max, so every request pays a full reasoning trace at $15/M output. Higher rate plus no way to dial down effort means K3's effective cost-per-task gap over GLM-5.2 is wider than the 3x sticker.

The one place K3's pricing genuinely wins is flat 1M context. Anthropic charges a premium above 200K input tokens and Google tiers Gemini by context length; Moonshot charges $3/M whether you send 4K or 900K tokens. For repo-scale analysis or long agent traces, that flatness can matter more than the headline rate. GLM-5.2 also prices its 1M window without a long-context tier, at a lower base rate, so for pure long-context economics GLM-5.2 is still cheaper per token; K3's flatness only pulls ahead if you need its specific capabilities in that window.

Verifiability and Open Weights

This is the axis that separates them most cleanly today, and it is the one that decides the "best open-source coding model" question. GLM-5.2 is open in the full sense right now: weights on Hugging Face (zai-org/GLM-5.2) under MIT with no regional restrictions, and its published scores have independent replication. You can download it, fine-tune it, serve it on your own hardware, and audit its behavior.

Kimi K3 launched API-only. As of July 16 there was no model card, license file, or download, and every K3 number was Moonshot-reported. What changed within a day is that Moonshot committed to an open-weights release on July 27 under a Modified MIT license, following the K2 family (K2, K2.5, K2.6, and K2-Thinking all shipped open). So the verifiability gap is real but closing: independent benchmark scores already landed, and downloadable weights are days out rather than a Q4 speculation.

Vision, Context, and Self-Host

Three capability differences decide many real choices before benchmarks enter the picture:

• Vision. K3 has native multimodal input for images and video; GLM-5.2 is text-only. FastAI's Jeremy Howard, who otherwise rated GLM-5.2 "at least as good as Opus 4.8 and GPT-5.5" for his coding work, flagged the missing vision as its main gap. If you feed the model screenshots, diagrams, or design mockups, K3 is the only option here.
• Context. Both run 1M tokens. The difference is what you fill it with. GLM-5.2 is verbose enough that 37K reasoning tokens per turn can crowd out your code in a long agent run, so context management pays off. K3's KDA design cuts KV-cache memory up to 75%, which is why its 1M window is priced flat, but its always-on max reasoning fills that window fast too.
• Self-host. GLM-5.2 today; K3 after July 27, at a much heavier footprint (see above). For teams that need on-prem or air-gapped deployment now, GLM-5.2 is the only one of the two that exists as a downloadable model.

Which One to Pick, by Workload

A common practical answer is to route: run GLM-5.2 as the high-volume default and reach for K3 on the hard, multimodal, or long-horizon tasks where its capability edge is worth 5x the token bill. That is exactly the pattern the Morph model lineup is built to serve, one OpenAI-compatible API across models, priced per token.

Running GLM-5.2 on Morph

Morph serves GLM-5.2 as morph-glm52-744b at $1.10/M input and $4.10/M output with the full 1M context, on custom codegen kernels with speculators trained on coding traffic. That is below Z.ai's own list price, and it matters because serving fidelity diverges across hosts: many serverless providers quantize activations to FP8 to cut cost, which moves output away from the reference weights. Kimi K3 is served only by Moonshot (and routed on OpenRouter from one upstream provider at launch); Morph does not serve it.

from openai import OpenAI

client = OpenAI(
    base_url="https://api.morphllm.com/v1",
    api_key="YOUR_MORPH_API_KEY",  # Authorization: Bearer <key>
)

resp = client.chat.completions.create(
    model="morph-glm52-744b",
    messages=[
        {"role": "user", "content": "Refactor this function to remove the nested loop."},
    ],
)
print(resp.choices[0].message.content)

The same key works for every model in the Open Source Models lineup, including MiniMax M3, Qwen 3.5 397B, and DeepSeek V4 Flash. For K3 access, use Moonshot's API at https://api.moonshot.ai/v1 with model kimi-k3; see the Kimi API guide.

GLM-5.2 vs Kimi K3: The Trade-Offs

Frequently Asked Questions

Is Kimi K3 better than GLM-5.2 for coding?

On capability, yes, by a modest margin. K3 leads vendor-reported agentic benchmarks (81.2 FrontierSWE, 88.3 Terminal-Bench 2.0) and the independent Artificial Analysis Intelligence Index (57 vs 51). But on the one apples-to-apples independent number, Terminal-Bench 2.1, they are effectively tied (K3 80.9 via Vals, GLM-5.2 81.0). GLM-5.2 wins on price (about one-third the output cost), open weights you can hold today, and independent replication of its scores.

What is the price difference between GLM-5.2 and Kimi K3?

K3 is roughly 3x on output: Moonshot charges $3/M input ($0.30 cached) and $15/M output; Z.ai lists GLM-5.2 at $1.40/$4.40, and Morph serves it at $1.10/$4.10. K3's always-on max reasoning widens the effective per-task gap beyond the 3x sticker, since you cannot cap its effort at launch.

Can I self-host GLM-5.2 and Kimi K3?

GLM-5.2, yes, today: MIT weights on Hugging Face, ~1.5TB BF16, 8x H200/B200 for a single node. K3, not at launch: API-only on July 16, weights promised July 27 under Modified MIT. At 2.8T total, K3 is ~5.6TB BF16, so even after weights land, self-hosting needs multi-node clusters or quantized variants.

Does GLM-5.2 or Kimi K3 have vision?

Only K3. It has native vision for images and video. GLM-5.2 is text-only, a gap Jeremy Howard flagged in his otherwise positive review.

Which is the best open-source coding model in 2026?

For weights you can download and run today, GLM-5.2: MIT, on Hugging Face, independently replicated. K3 scores higher and adds vision, but is API-only until July 27. After the weights ship, K3 becomes the higher-capability open-weights option at a much heavier serving footprint and a higher hosted price. See Best Open-Source Coding Models in 2026.

How many parameters do GLM-5.2 and Kimi K3 have?

GLM-5.2 is 753B total, ~40B active per token. K3 is 2.8T total, ~50B active (estimated; Moonshot has not officially disclosed the active count). Total drives memory and self-host cost; active drives per-token compute.