LLM Pareto Frontier — price vs intelligence, updated monthly
A model is Pareto-efficient if no other model is both cheaper and smarter. This report plots the current flagship LLMs — the open-weight flagships (including every one we serve) and the closed models from Anthropic, OpenAI and Google — on output price versus the Artificial Analysis Intelligence Index, and computes the frontier from the data. Everything below the line is dominated: a strictly better deal exists.
Method: all numbers come from the same source on the same day — Artificial Analysis model pages (Intelligence Index, vendor list prices). Each edition is archived so you can watch the frontier move over time.
Current frontier (edition 2026-10)
Section titled “Current frontier (edition 2026-10)”Reading left to right: MiMo-V2.6-Flash → GLM-5.3-Flash → MiMo-V2.6-Pro → Claude Sonnet 5.5 → Claude Opus 5.5.
Analysis
Section titled “Analysis”- Kimi K3.1 Pro is now plotted as open-weights (2026-11-05): index 55 at $2/$10 — #4 of everything AA tracks, 3 points off the leader — and Moonshot published the weights on November 4 — the day it had promised — under a license that permits commercial serving. It is the top open-weights score on the current index (v4.3.2), 9 points above the previous best (MiMo-V2.6-Pro, 46). It still sits one point under the frontier line, not on it: Claude Sonnet 5.5 scores 56 at the identical $10/M. It dominates Kimi K3 outright (44 at $15), GPT-6.1 Sol, GPT-5.4 and every closed model priced above it except the two Anthropic 5.5 points. Full analysis: Kimi K3.1 Pro: specs, benchmarks, pricing — since 2026-11-06 it is served in our Flagship Pool.
- Three closed models join to keep the top of the chart honest: Claude Opus 5.5 (58, $4/$20), Claude Sonnet 5.5 (56, $2/$10) and GPT-6.1 Sol (52, $2/$10). Opus 5.5 is the new index leader and the new top of the frontier — at less than half the output price of GPT-6 Astra and Claude Fable 5.1 (53 at $50), which it dominates along with Claude Opus 5 and GPT-5.6 Sol. Sonnet 5.5 takes the step below it and dominates GPT-6.1 Sol at the same price. AA now marks Opus 5, GPT-5.6 Sol and Kimi K3 as superseded; they stay tracked for history.
- The frontier reads MiMo-V2.6-Flash (38, $0.28) → GLM-5.3-Flash (42, $0.50) → MiMo-V2.6-Pro (46, $0.87) → Claude Sonnet 5.5 (56, $10) → Claude Opus 5.5 (58, $20). The closed half of the line collapsed from four points to two, both Anthropic’s; the open half is unchanged.
- The open-vs-closed gap is 3 points on today’s chart (Kimi K3.1 Pro 55 vs Claude Opus 5.5 58) — it was 12 (MiMo-V2.6-Pro 46) until K3.1 Pro’s weights landed on 2026-11-04, the narrowest gap in this report’s history. The step from the best open model to the next closed one is now 1 point at the same output price (K3.1 Pro → Sonnet 5.5, both $10); before the weights it was 10 points for 11.5× the price (MiMo-V2.6-Pro $0.87 → Sonnet 5.5 $10).
- The DeepSeek V4.1 Flash point now tracks the served 1017 build on its own AA page: 45 at $1.20/M, +6 at the same price — it matches GLM 5.3 at 3.7× less and Qwen3.8 Max at 5× less, and is still dominated only by MiMo-V2.6-Pro.
- Every tracked model priced under $5 per million output tokens is open-weight. GLM-5.3-Flash still anchors the cheap end ($0.50, 42) and matches Claude Opus 4.8 at 50× less. GLM 5.3, Kimi K3.1 Pro, Qwen3.8 Max, DeepSeek V4.1 Flash and MiMo V2.6 Flash are served in our pools (Kimi K3.1 Pro in the Flagship Pool since 2026-11-06, replacing Kimi K3 in place); MiMo-V2.6-Pro and GLM-5.3-Flash are under evaluation (live status).
- Scores on this chart are Intelligence Index v4.3.2 (since 2026-09-22): a harder composite than the v4.1.1 used through August, which rescored the whole field 12–19 points lower at unchanged prices. Scores from earlier editions are not comparable; compare positions, which is what the frontier is for.
The data
Section titled “The data”| Model | AA Index | Input $/M | Output $/M | Weights | Pareto-efficient |
|---|---|---|---|---|---|
| Claude Opus 5.5 | 58 | $4.00 | $20.00 | closed | ✓ |
| Claude Sonnet 5.5 | 56 | $2.00 | $10.00 | closed | ✓ |
| Kimi K3.1 Pro | 55 | $2.00 | $10.00 | open | dominated by Claude Sonnet 5.5 |
| GPT-6 Astra | 53 | $10.00 | $50.00 | closed | dominated by Claude Opus 5.5 |
| Claude Fable 5.1 | 53 | $10.00 | $50.00 | closed | dominated by Claude Opus 5.5 |
| GPT-6.1 Sol | 52 | $2.00 | $10.00 | closed | dominated by Claude Sonnet 5.5 |
| Claude Opus 5 | 51 | $5.00 | $25.00 | closed | dominated by Claude Opus 5.5 |
| Claude Fable 5 | 50 | $10.00 | $50.00 | closed | dominated by Claude Opus 5.5 |
| GPT-5.6 Sol | 47 | $4.00 | $20.00 | closed | dominated by Claude Opus 5.5 |
| MiMo-V2.6-Pro | 46 | $0.43 | $0.87 | open | ✓ |
| DeepSeek V4.1 Flash | 45 | $0.30 | $1.20 | open | dominated by MiMo-V2.6-Pro |
| GLM 5.3 | 45 | $1.40 | $4.40 | open | matched by DeepSeek V4.1 Flash at 3.7× less |
| Qwen3.8 Max | 45 | $2.00 | $6.00 | open | matched by DeepSeek V4.1 Flash at 5.0× less |
| Kimi K3 | 44 | $3.00 | $15.00 | open | dominated by Claude Sonnet 5.5 |
| GLM-5.3-Flash | 42 | $0.15 | $0.50 | open | ✓ |
| Opus 4.8 | 42 | $5.00 | $25.00 | closed | matched by GLM-5.3-Flash at 50.0× less |
| GPT-5.4 | 39 | $2.50 | $15.00 | closed | dominated by Claude Sonnet 5.5 |
| MiMo-V2.6-Flash | 38 | $0.14 | $0.28 | open | ✓ |
| Sonnet 5 | 38 | $2.00 | $10.00 | closed | matched by MiMo-V2.6-Flash at 35.7× less |
| GPT-5.5 | 38 | $5.00 | $30.00 | closed | matched by MiMo-V2.6-Flash at 107.1× less |
| DeepSeek V4 Flash | 34 | $0.14 | $0.28 | open | dominated by MiMo-V2.6-Flash |
| GLM 5.2 | 34 | $1.40 | $4.40 | open | matched by DeepSeek V4 Flash at 15.7× less |
| Gemini 3.5 Flash | 33 | $1.50 | $9.00 | closed | dominated by MiMo-V2.6-Pro |
| MiniMax M3 | 29 | $0.30 | $1.20 | open | dominated by MiMo-V2.6-Pro |
| Kimi K2.6 | 27 | $0.95 | $4.00 | open | dominated by MiMo-V2.6-Pro |
| Kimi K2.7 | 26 | $0.95 | $4.00 | open | dominated by MiMo-V2.6-Pro |
How the frontier moves
Section titled “How the frontier moves”Each edition’s frontier is archived; this chart overlays them, current on top. Over time it shows the defining dynamic of this market: the frontier sliding down (cheaper) and right-side-up (smarter) — driven almost entirely by open-weight releases.
Caveats
Section titled “Caveats”The Intelligence Index is one composite — task-specific rankings differ. Kimi K2.6 sits off-frontier here yet holds the best open SWE-bench Verified score (80.2); for agentic coding the ranking flips — see the Which-LLM guide for tier-fair matchups. Prices are vendor list prices for the reasoning variants Artificial Analysis evaluates; open-weight prices vary by host. Kimi K3.1 Pro was plotted as closed-weights from 2026-10-26 until Moonshot published the weights on 2026-11-04; its point and row flipped to open on 2026-11-05 with no number changed. Index points aren’t linear in value.
Changelog
Section titled “Changelog”-
2026-10 (updated 2026-11-05, no frontier change) — Kimi K3.1 Pro flips to open-weights: Moonshot published the weights on November 4 under the Kimi K3.1-Pro License (commercial serving permitted). No score or price moved and the frontier is unchanged — K3.1 Pro stays one point under Claude Sonnet 5.5 at the same $10/M — but the chart’s open-vs-closed gap drops from 12 points to 3, and the best open score on the current index goes from 46 (MiMo-V2.6-Pro) to 55. Still confirmed for our Flagship Pool, pending capacity validation.
-
2026-10 (updated 2026-10-26) — Kimi K3.1 Pro joins the tracked set (55, $2/$10): Moonshot’s post-trained K3, #4 on the index and 3 points off the leader, plotted as closed until the weights announced for 2026-11-04 are published — it is confirmed for our Flagship Pool. Claude Opus 5.5 (58, $4/$20), Claude Sonnet 5.5 (56, $2/$10) and GPT-6.1 Sol (52, $2/$10) join as well: Opus 5.5 is the new index leader and the new top of the frontier; Sonnet 5.5 edges K3.1 Pro by a point at the same price and takes the step below; GPT-6 Astra, Claude Fable 5.1, Claude Opus 5 and GPT-5.6 Sol all leave the line, dominated. The DeepSeek V4.1 Flash point now follows the served 1017 build on its own AA page (45, +6 at the same price). Frontier: MiMo-V2.6-Flash → GLM-5.3-Flash → MiMo-V2.6-Pro → Claude Sonnet 5.5 → Claude Opus 5.5.
-
2026-10 (updated 2026-10-18) — a quiet month for scores: every previously tracked model is unchanged in index. MiMo V2.6 Flash gets its Artificial Analysis entry, joins the tracked set (38, $0.14/$0.28, MIT weights) and enters the frontier as its new low-price anchor — nothing scored 38 or below comes close on price. Frontier: MiMo-V2.6-Flash → GLM-5.3-Flash → MiMo-V2.6-Pro → GPT-5.6 Sol → Claude Opus 5 → GPT-6 Astra → Claude Fable 5.1. AA also re-lists the retired DeepSeek V4 Flash at $0.14/$0.28 (was $0.44/$1.32). DeepSeek shipped V4.1-Flash-1017 on 2026-10-17 (what changed) — AA has not re-scored it, so the V4.1 Flash point still reflects the September build.
-
2026-09-23 (no frontier change) — MiMo V2.6 Flash replaces MiMo V2.5 in our Core Pool in place, at the same $0.14/$0.28 list price: 309B MoE / 15B active, image input, MIT weights. Artificial Analysis has not scored it, so it stays on the watchlist and no point on this chart moves.
-
2026-09 (2026-09-22) — AA moved its index to v4.3.2: every previously tracked score fell 12–19 points at unchanged prices — cross-edition drops at this date are methodology, not model changes. MiMo-V2.6-Pro joins the tracked set (46, $0.43/$0.87, MIT weights) as the #1 open-weights model and enters the frontier, dominating GLM 5.3, Kimi K3, Sonnet 5 and DeepSeek V4.1 Flash — GLM 5.3 leaves the line. DeepSeek V4.1 Flash is scored (39, $0.30/$1.20 peak) and joins the tracked set. GPT-6 Astra and Claude Fable 5.1 (both 53, $10/$50) join the tracked set as the new top of the index; Claude Fable 5 becomes dominated. Frontier: GLM-5.3-Flash → MiMo-V2.6-Pro → GPT-5.6 Sol → Claude Opus 5 → GPT-6 Astra / Claude Fable 5.1. Analysis: MiMo-V2.6 Pro & Flash.
-
2026-09-10 (no frontier change) — DeepSeek V4.1 Flash ships and replaces V4 Flash in our Core Pool: 552B MoE (8B/16B active), causal encoder–decoder, native vision, MIT weights, $0.30/$1.20 peak list — cheaper than the model it replaces. Artificial Analysis has not scored it, so it is added to the watchlist and no point on this chart moves; the DeepSeek point stays V4 Flash until an independent index exists.
-
2026-08 refresh (2026-08-31) — GLM-5.3-Flash joins the tracked set (57, $0.15/$0.50) and lands straight on the frontier: Z.ai’s 18B-active, MIT-licensed, natively multimodal MoE matches Claude Opus 4.8 at 50× less and dominates DeepSeek V4 Flash, MiniMax M3 and Kimi K2.7 — the frontier’s whole cheap half collapses into one point. Frontier: GLM-5.3-Flash → GLM 5.3 → GPT-5.6 Sol → Claude Opus 5. Analysis: GLM-5.3-Flash: specs, benchmarks & pricing.
-
2026-08 refresh (2026-08-30) — GLM 5.3 joins the tracked set (60, $1.40/$4.40) two days after its open weights shipped under the GLM-5.3 License, and enters the frontier at once: it dominates Qwen3.8 Max, Sonnet 5 and GLM 5.2, and matches Kimi K3 at 3.4× less — K3 leaves the line. GPT-5.6 Sol’s output price drops $30 → $20 and it re-enters the frontier. Frontier: MiniMax M3 → DeepSeek V4 Flash → GLM 5.3 → GPT-5.6 Sol → Claude Opus 5. GLM 5.2 stays tracked for history; it was upgraded in place to 5.3 in our Frontier Pool.
-
2026-08 refresh (2026-08-14) — three moves in one collection. AA recalibrated its index to v4.1.1: every tracked score shifted up 1–5 points at unchanged prices — cross-edition score jumps around this date are methodology, not model changes. Qwen3.8 Max joins the tracked set (58, $2/$6) now that its weights shipped — it enters the frontier and pushes Sonnet 5 off it, leaving Opus 5 as the only closed model on the line. And DeepSeek V4 Flash’s peak list pricing took effect ($0.14/$0.28 → $0.44/$1.32), bringing MiniMax M3 back onto the frontier. Frontier: MiniMax M3 → DeepSeek V4 Flash → GLM 5.2 → Qwen3.8 Max → Kimi K3 → Claude Opus 5.
-
2026-08 (2026-08-01) — release-week update: the V4-Flash-0731 retrain lifts DeepSeek V4 Flash 40 → 50 at unchanged prices ($0.14/$0.28) — a 10-point single-model jump. It now matches Gemini 3.5 Flash at 32× lower output price and pushes MiniMax M3 off the frontier (dominated). Frontier: DeepSeek V4 Flash → GLM 5.2 → Sonnet 5 → Kimi K3 → Claude Opus 5.
-
2026-07 refresh (2026-07-28) — release-week update: Kimi K3 (57, $3/$15), Claude Opus 5 (61, $5/$25) and GPT-5.6 Sol (59, $5/$30) added to the tracked set. K3 and Opus 5 enter the frontier; Fable 5 and Opus 4.8 leave it (both dominated by Opus 5). Sonnet 5’s output price dropped $15 → $10. The open-vs-closed gap is now 4 index points — the narrowest since GLM-5 (February).
-
2026-07 (first edition) — baseline. Frontier: DeepSeek V4 Flash, MiniMax M3, GLM 5.2, Sonnet 5, Opus 4.8, Fable 5. Notable context at launch: GLM 5.2 (released mid-June) is the top-scoring open-weights model in index history; AA’s v4.1 recalibration lowered scores across the board vs the April v4.0 figures.
GLM 5.3, Kimi K3.1 Pro, Qwen3.8 Max, Kimi K3, DeepSeek V4.1 Flash and MiMo V2.6 Flash are served flat-rate in our pools — Flagship from $169.15/mo, Frontier from $60.35/mo, Core from $18.70/mo — where the marginal token costs zero during your reserved hours. Kimi K3.1 Pro joined the Flagship Pool on 2026-11-06; MiMo-V2.6-Pro and GLM-5.3-Flash are under evaluation.